WO2024019121A1

WO2024019121A1 - Motor driving system

Info

Publication number: WO2024019121A1
Application number: PCT/JP2023/026631
Authority: WO
Inventors: 敬介藤▲崎▼; 賢哉成瀬
Original assignee: 学校法人トヨタ学園
Priority date: 2022-07-21
Filing date: 2023-07-20
Publication date: 2024-01-25
Also published as: JP2024014146A

Abstract

[Problem] To provide a motor driving system of which loss is reduced further than a case where a sinusoidal wave, which is a fundamental wave, is adopted as a signal wave. [Solution] This motor driving system 1 comprises: a 3-phase inverter unit 2 which outputs a 3-phased PWM driving voltage that drives a permanent magnet synchronization motor 5; and a motor control unit 4 which controls the 3-phase inverter unit 2. This motor control unit 4 switches a pulse width at an intersection point of a carrier wave and a signal wave h(t)(=m•sin(2πf1₁t)+n•sin(2π•5 f₁t+φ)) in which a fifth-order harmonic wave is overlapped with a fundamental sinusoidal wave g(t) and generates a PWM drive signal, supplies the PWM drive signal to the 3-phase inverter unit 2 and causes a PWM driving voltage to be output. This motor driving system 1 operates with an upper limit of an overlap ratio n/m of the signal wave h(t) and a lower limit of the overlap ratio n/m, the upper limit being a value of n/m at which a fundamental wave current I_f1 begins to decrease due to a change in magnetic characteristics, and the lower limit being a value of n/m at which a decrease effect of the fundamental wave current I_f1 is greater than an increase due to harmonic components.

Description

motor drive system

The present invention relates to a motor drive system that drives a synchronous motor using a pulse width modulation control method.

Synchronous motors that exhibit high efficiency are widely used, and in particular, permanent magnet synchronous motors (PMSM) have been put into practical use as energy-saving, high-performance motors for electric vehicles, air conditioners, refrigerators, etc. is progressing. A PWM control method is often adopted to drive a synchronous motor, in which a pulse width modulation (PWM) voltage supplied from an inverter circuit is input and controlled. A PWM drive signal for forming a PWM voltage is input to an inverter circuit that outputs a PWM voltage. In this PWM drive signal, the signal wave (modulation wave) is a sine wave signal, the carrier wave (carrier wave) is a triangular wave signal, and the pulse width is determined so that the intersection of the signal wave and the carrier wave becomes an edge. . Various proposals have been made for the method of generating this PWM drive signal. For example, in Patent Document 1 listed below, a waveform obtained by adding and superimposing harmonic components to a sine wave signal serving as a fundamental wave is used as a signal wave. A method is disclosed in which the pulse width is determined based on the intersection of the signal wave and the carrier wave. According to this PWM signal generation method, it is stated that the motor output capacity is large and the synchronous motor can be driven with low vibration and low noise.

Japanese Patent Application Publication No. 2012-135100

However, when harmonic components are added to the signal wave as in the method described in Patent Document 1, the effective value increases due to the harmonic components, and it is generally assumed that loss increases. This is as expected if it is within the range of linear characteristics, but if it has magnetic nonlinearity, such as the magnetic material used in the motor core, it is expected that the loss may be reduced by superimposing harmonics. .

Therefore, an object of the present invention is to provide a motor drive system with reduced loss compared to the case where a sine wave serving as a fundamental wave is used as a signal wave.

In order to solve this problem, the invention according to claim 1 is a motor drive system that drives a motor with an inverter, which generates a signal wave h(t) obtained by superimposing a harmonic on a fundamental sine wave g(t). In the switching operation of the semiconductor in the inverter at the intersection of carrier waves, the fundamental sine wave g(t) is g(t)=m·sin(2πf ₁ t), and the signal wave h(t) is h( t)=m・sin(2πf ₁ t)+n・sin(2π・5f ₁ t+φ), and the upper limit of n/m is the value of n/m at which the fundamental wave current starts to decrease due to changes in magnetic properties, n The present invention is characterized in that the lower limit of /m is a value of n/m at which the increase due to harmonic components is greater than the reduction effect of the fundamental wave current.

The invention according to claim 2 is characterized in that, in addition to the configuration according to claim 1, the invention operates with n/m greater than −0.3 and less than 0.

The invention according to claim 3 provides, in addition to the configuration according to claim 1 or 2, when the fundamental sine wave g(t) has a phase angle (2πf ₁ t) of π/2 radian. of the harmonics in which the value m・sin(π/2) of (t) is greater than or equal to the value of the signal wave h(t)=m・sin(π/2)+n・sin(5・π/2+φ) It is characterized by operating with an initial phase φ.

In addition to the structure according to claim 1 or 2, the invention according to claim 4 is such that the maximum phase angle of the initial phase φ of the harmonic is the initial phase angle at which the fundamental wave current starts to decrease due to a change in the magnetic property. The operation is characterized in that the value of the phase φ and the minimum phase angle of the initial phase φ are such that the increase due to the harmonic component is greater than the reduction effect of the fundamental wave current.

The invention according to claim 5 is a motor drive system in which a motor is driven by an inverter, in which a signal wave h(t) obtained by superimposing a harmonic wave on a fundamental sine wave g(t) and a carrier wave intersect within the inverter. In the _{semiconductor} switching operation of ₁ t) + n・sin (2π・af ₁ t+φ), where a is an integer of 5 or more, and the upper limit of n/m is the value of n/m at which the fundamental wave current starts to decrease due to changes in magnetic properties. , n/m is operated at a value where the lower limit of n/m is such that the increase due to harmonic components is greater than the reduction effect of the fundamental wave current.

The invention according to claim 6 is characterized in that, in addition to the configuration according to claim 5, the invention operates with n/m greater than −0.3 and less than 0.

The invention according to claim 7 provides, in addition to the configuration according to claim 5 or 6, when the fundamental sine wave g(t) has a phase angle (2πf ₁ t) of π/2 radian. The harmonic wave whose numerical value m・sin(π/2) of (t) is greater than or equal to the value of the signal wave h(t)=m・sin(π/2)+n・sin(a・π/2+φ) It is characterized by operating with an initial phase φ.

In addition to the configuration described in claim 5 or 6, the invention according to claim 8 provides that the maximum phase angle of the initial phase φ of the harmonic is the initial phase angle at which the fundamental wave current starts to decrease due to a change in the magnetic property. The operation is characterized in that the value of the phase φ and the minimum phase angle of the initial phase φ are such that the increase due to the harmonic component is greater than the reduction effect of the fundamental wave current.

The invention according to claim 9 provides a motor comprising: an inverter section that outputs a pulse width modulated drive voltage for driving a synchronous motor; and a motor control section that controls the inverter section so as to form the pulse width modulation drive voltage. In the drive system, the motor control unit generates a pulse at the intersection of a carrier wave and a signal wave obtained by superimposing a fifth harmonic of the fundamental sine wave on a fundamental sine wave that defines the fundamental frequency of the pulse width modulated drive voltage. It is configured to generate a PWM drive signal that changes the width to form the pulse width modulated drive voltage, and supplies the PWM drive signal to the switching element input section of the inverter section, and to change the frequency of the fundamental sine wave. When the fundamental sine wave frequency f is ₁ , the signal wave h(t) is set as h(t)=m・sin(2πf ₁ t)+n・sin(2π・5f ₁ t+φ), and the modulation rate of the fundamental sine wave is The upper and lower limits of the superimposition rate n/m, which is the ratio of the modulation rate n of the fifth harmonic of the initial phase φ to m, are the primary A pulse width modulated voltage using the signal wave h(t) is applied to the coil while changing the superimposition ratio n/m, and the primary current flowing through the primary coil and the secondary coil wound around the ring sample are The upper limit value of the superimposition ratio n/m is determined by a ring test that measures the secondary voltage generated in the first harmonic than when the modulation ratio n of the fifth harmonic is zero. The superposition ratio n/m, which is the upper limit of the range below which the fundamental wave current, which is the component of the fundamental sine wave frequency _f1 of the current, is zero, or the modulation rate n of the fifth harmonic is zero. The superimposition ratio n/m is determined as the upper limit of the range below which the predetermined loss calculated based on the primary current and the secondary voltage falls, and the lower limit value of the superimposition ratio n/m is the fifth harmonic. The fundamental sine wave frequency of the primary current is based on the case where the modulation rate n of the fifth harmonic is zero, compared to the amount of reduction of the fundamental wave current based on the case where the modulation rate n of the fifth harmonic is zero. The predetermined loss is greater than when the superimposition ratio n/m, which is the lower limit of the range below which the amount of increase in the harmonic current, which is the harmonic component of _f1 , or the modulation ratio n of the fifth harmonic is zero. The superimposition ratio is determined as n/m, which is the lower limit of the range below.

In the invention according to claim 10, in addition to the configuration according to claim 9, the maximum phase angle and the minimum phase angle of the initial phase φ of the fifth harmonic are changed by changing the initial phase φ to adjust the pulse width. The maximum phase angle is determined by the ring test in which a modulation voltage is applied to the primary coil, and the maximum phase angle is within a range in which the fundamental current is lower than when the modulation rate n of the fifth harmonic is zero. The initial phase φ is determined as the maximum value of φ, or the initial phase φ is determined as the maximum value of the range in which the predetermined loss is lower than when the modulation rate n of the fifth harmonic is zero, and the minimum phase The angle is the amount of reduction in the fundamental wave current based on the case where the modulation rate n of the fifth-order harmonic is zero, compared to the amount of reduction of the fundamental wave current based on the case where the modulation rate n of the fifth-order harmonic is zero. The initial phase φ is a minimum value in a range where the increase in wave current is below, or the initial phase is a minimum value in a range where the predetermined loss is lower than when the modulation rate n of the fifth harmonic is zero. It is characterized in that it is determined as φ.

The invention according to claim 11 provides a motor comprising: an inverter section that outputs a pulse width modulated drive voltage for driving a synchronous motor; and a motor control section that controls the inverter section so as to form the pulse width modulation drive voltage. In the drive system, the motor control unit generates a pulse at the intersection of a carrier wave and a signal wave obtained by superimposing a fifth harmonic of the fundamental sine wave on a fundamental sine wave that defines the fundamental frequency of the pulse width modulated drive voltage. It is configured to generate a PWM drive signal that changes the width to form the pulse width modulated drive voltage, and supplies the PWM drive signal to the switching element input section of the inverter section, and to change the frequency of the fundamental sine wave. When the fundamental sine wave frequency f is ₁ , the signal wave h(t) is set as h(t)=m・sin(2πf ₁ t)+n・sin(2π・5f ₁ t+φ), and the modulation rate of the fundamental sine wave is The upper and lower limits of the superimposition rate n/m, which is the ratio of the modulation rate n of the fifth harmonic of the initial phase φ to m, are determined by the pulse width modulation drive voltage using the signal wave h(t). The synchronous motor is rotationally driven by changing the superimposition ratio n/m, and at least one of the input power to the inverter section, the input power of each phase of the synchronous motor, or the input current of each phase of the synchronous motor is measured. The upper limit of the superimposition ratio n/m is determined by a motor test in which the fundamental wave current, which is a component of the fundamental sine wave frequency _f1 calculated from the input current, is The superimposition ratio n/m is the upper limit of the range below when the harmonic modulation ratio n is zero, or the input power to the inverter section, the input power to each phase of the synchronous motor, or the input power to each phase of the synchronous motor. The predetermined loss calculated based on at least one of the input currents is determined as the superimposition ratio n/m that is the upper limit of a range in which the modulation ratio n of the fifth harmonic is lower than when the modulation ratio n is zero, The lower limit value of the superposition ratio n/m is such that the modulation rate n of the fifth harmonic is zero compared to the amount of reduction in the fundamental wave current based on the case where the modulation rate n of the fifth harmonic is zero. The superimposition ratio n/m is the lower limit of the range below which the amount of increase in the harmonic current, which is a harmonic component of the fundamental sine wave frequency _f1 of the input current, is based on the case where It is characterized in that the superimposition ratio n/m is determined as the lower limit of the range in which the predetermined loss is lower than when the wave modulation ratio n is zero.

In the invention according to claim 12, in addition to the configuration according to claim 11, the maximum phase angle and the minimum phase angle of the initial phase φ of the fifth harmonic are changed by changing the initial phase φ to adjust the pulse width. The maximum phase angle is determined by the motor test in which the synchronous motor is rotationally driven by a modulated drive voltage, and the maximum phase angle is determined when the fundamental wave current is lower than when the modulation rate n of the fifth harmonic is zero. The initial phase φ is determined as the maximum value in the range below which the predetermined loss falls below that when the modulation rate n of the fifth harmonic is zero, and The minimum phase angle is based on the case where the modulation rate n of the fifth-order harmonic is zero compared to the amount of reduction in the fundamental wave current based on the case where the modulation rate n of the fifth-order harmonic is zero. The initial phase φ is a minimum value in a range below which the increase amount of the harmonic current is lower, or the initial phase φ is a minimum value in a range where the predetermined loss is lower than when the modulation rate n of the fifth harmonic is zero. It is characterized in that it is determined as the initial phase φ.

According to the invention of claim 1, the fifth harmonic is superimposed on the signal wave, and the upper limit of the superimposition ratio n/m is the value of n/m at which the fundamental wave current starts to decrease due to a change in magnetic characteristics, and n/m is the upper limit of the superimposition ratio n/m. It operates so that the lower limit of m is a value of n/m at which the increase due to harmonic components is greater than the effect of reducing the fundamental wave current. Because of this, the loss of the motor drive system can be reduced compared to the case where a fundamental sine wave is used as a signal wave.

According to the invention of claim 2, since the superimposition ratio n/m operates within a range of greater than -0.3 and less than 0, it is possible to stably reduce the loss of the motor drive system.

According to the invention of claim 3, the numerical value m _· sin(π/ 2) operates with an initial phase φ of the harmonic that is greater than or equal to the value of the signal wave h(t)=m·sin(π/2)+n·sin(5·π/2+φ). Because of this, the loss of the motor drive system can be reduced compared to the case where the fundamental sine wave g(t) is used as the signal wave h(t).

According to the invention of claim 4, the maximum phase angle of the initial phase φ of the harmonics is the value of the initial phase φ at which the fundamental wave current starts to decrease due to a change in magnetic characteristics, and the minimum phase angle of the initial phase φ is the value of the fundamental wave current. It operates at an initial phase φ value at which the increase due to harmonic components is greater than the current reduction effect. Because of this, the loss of the motor drive system can be reduced compared to the case where a fundamental sine wave is used as a signal wave.

According to the invention of claim 5, the fifth or higher harmonics are superimposed on the signal wave, and the upper limit of the superimposition ratio n/m is the value n/m at which the fundamental wave current starts to decrease due to a change in magnetic characteristics. The lower limit of n/m is a value of n/m at which the increase due to harmonic components is greater than the effect of reducing the fundamental wave current. Because of this, the loss of the motor drive system can be reduced compared to the case where a fundamental sine wave is used as a signal wave.

According to the invention of claim 6, since the superimposition ratio n/m operates within a range of greater than −0.3 and less than 0, it is possible to stably reduce loss.

According to the invention of claim 7, when the phase angle (2πf ₁ t) of the fundamental sine wave g(t) is π/2 radian, the numerical value m·sin(π/2) of the fundamental sine wave g(t) is , the signal wave operates at an initial phase φ of the harmonic that is greater than or equal to the value of h(t)=m·sin(π/2)+n·sin(a·π/2+φ). Because of this, loss can be reduced compared to the case where the fundamental sine wave g(t) is used as the signal wave h(t).

According to the invention of claim 8, the maximum phase angle of the initial phase φ of the harmonic is the value of the initial phase φ at which the fundamental wave current starts to decrease due to a change in magnetic characteristics, and the minimum phase angle of the initial phase φ is It operates at an initial phase φ value at which the increase due to harmonic components is greater than the current reduction effect. Because of this, loss can be reduced compared to the case where a fundamental sine wave is used as a signal wave.

According to the invention of claim 9, the upper and lower limits of the superimposition rate n/m, which is the ratio of the modulation rate n of the fifth harmonic to the modulation rate m of the fundamental sine wave constituting the signal wave, are determined by a ring test. The upper limit of the superposition ratio n/m is the upper limit of the range in which the fundamental wave current is lower than when the modulation rate n of the fifth harmonic is zero, or The modulation rate n/m is determined as the upper limit of the range where the predetermined loss is lower than when the modulation rate n of the 5th harmonic is zero, and the lower limit of the modulation rate n/m is the modulation rate of the 5th harmonic. Superposition that is the lower limit of the range in which the amount of increase in harmonic current based on the case where the modulation rate n of the fifth harmonic is zero is lower than the amount of reduction in the fundamental wave current based on the case where the modulation rate n of the fifth harmonic is zero. The modulation rate n/m or the superimposition rate n/m is determined as the lower limit of the range in which the predetermined loss is lower than when the modulation rate n of the fifth harmonic is zero. Because of this, loss can be reduced compared to the case where a fundamental sine wave is used as a signal wave.

According to the invention of claim 10, the maximum phase angle and minimum phase angle of the initial phase φ of the fifth harmonic are determined by a ring test performed by changing the initial phase φ, and the maximum phase angle is the initial phase φ that is the maximum value in the range in which the fundamental wave current is lower than when the modulation rate n of the fifth harmonic is zero, or the predetermined loss than when the modulation rate n of the fifth harmonic is zero. The minimum phase angle is determined as the initial phase φ that is the maximum value in the range in which the 5th harmonic The initial phase φ is the minimum value in the range in which the increase in harmonic current is below the case where the modulation rate n of the fifth harmonic is zero, or the predetermined loss is lower than when the modulation rate n of the fifth harmonic is zero. The initial phase φ is set to be the minimum value within the range below. Because of this, loss can be reduced compared to the case where a fundamental sine wave is used as a signal wave.

According to the invention of claim 11, the upper and lower limits of the superimposition rate n/m, which is the ratio of the modulation rate n of the fifth harmonic to the modulation rate m of the fundamental sine wave constituting the signal wave, are determined by a motor test. The upper limit of the superposition ratio n/m is the superposition ratio n/m that is the upper limit of the range in which the fundamental wave current is lower than when the modulation rate n of the fifth harmonic is zero, or The upper limit of the range where the predetermined loss is lower than when the modulation rate n of the fifth harmonic is zero is determined as the superimposition rate n/m, and the lower limit value of the superimposition rate n/m is the modulation rate of the fifth harmonic. Superimposition rate that is the lower limit of the range in which the amount of increase in harmonic current based on the case where the modulation rate n of the fifth harmonic is zero is lower than the amount of reduction in the fundamental wave current based on the case where n is zero. n/m, or a superimposition rate n/m that is the lower limit of a range in which the predetermined loss is lower than when the modulation rate n of the fifth harmonic is zero. Because of this, loss can be reduced compared to the case where a fundamental sine wave is used as a signal wave.

According to the invention of claim 12, the maximum phase angle and the minimum phase angle of the initial phase φ of the fifth harmonic are determined by a motor test performed by changing the initial phase φ, and the maximum phase angle is the initial phase φ that is the maximum value in the range in which the fundamental wave current is lower than when the modulation rate n of the fifth harmonic is zero, or the predetermined loss than when the modulation rate n of the fifth harmonic is zero. The minimum phase angle is determined as the initial phase φ that is the maximum value in the range in which the 5th harmonic The initial phase φ is the minimum value in the range in which the increase in harmonic current is below the case where the modulation rate n of the fifth harmonic is zero, or the predetermined loss is lower than when the modulation rate n of the fifth harmonic is zero. The initial phase φ is set to be the minimum value within the range below. Because of this, loss can be reduced compared to the case where a fundamental sine wave is used as a signal wave.

1 is a schematic configuration diagram of a motor drive system according to an embodiment of the present invention. It is a schematic block diagram of the motor control part of the motor drive system based on the same embodiment. FIG. 3 is a diagram illustrating pulse width modulation of a PWM drive signal of the motor drive system according to the same embodiment, in which (a) is a diagram showing a signal wave and a carrier wave, (b) is a diagram showing a PWM drive signal, and (c ) is a diagram showing the PWM drive voltage. FIG. 3 is a diagram showing three-phase fundamental sine waves of the motor drive system according to the embodiment. It is a figure showing the flow of operation of a motor control part of a motor drive system concerning the same embodiment. FIG. 2 is a schematic configuration diagram of a ring testing device used for setting, designing, and manufacturing a motor drive system according to the same embodiment. 3 is a diagram illustrating an example of setting conditions for a ring test used in setting, designing, and manufacturing a motor drive system according to the same embodiment; FIG. FIG. It is a figure which shows the example of the time waveform of the magnetic field strength and magnetic flux density by the ring test used for the setting, design, and manufacturing of the motor drive system based on the same embodiment, (a) is when the superimposition ratio n/m is 0. (b) is a diagram showing the major loop component when the superposition ratio n/m is 0, (c) is a diagram showing the measured waveform when the superposition ratio n/m is -0.2. , (d) are diagrams showing major loop components when the superimposition ratio n/m is -0.2. It is a figure which shows the example of the BH curve of the magnetic field strength and magnetic flux density by the ring test used for the setting, design, and manufacturing of the motor drive system based on the same embodiment, (a) is when the superimposition ratio n/m is 0. (b) is a diagram showing the measurement results and major loop component when the superposition ratio n/m is -0.2, (c) is a diagram showing the measurement results and the major loop component when the superposition ratio n/m is 0. FIG. 3 is a diagram showing a major loop component of -0.2. It is a figure which shows the example of the fundamental wave current and fundamental sine wave modulation factor by the ring test used for the setting, design, and manufacture of the motor drive system based on the same embodiment. It is a figure showing an example of iron loss, major loop iron loss, carrier harmonic iron loss (minor loop iron loss) by a ring test used for setting, design, and manufacturing of a motor drive system concerning the embodiment, and (a) is a diagram showing the iron loss, and (b) is a diagram showing the rate of change in the iron loss when the superimposition ratio n/m is 0 as a reference. FIG. 3 is a diagram showing an example of iron loss and fundamental wave current obtained by a ring test used in setting, designing, and manufacturing the motor drive system according to the embodiment. It is a figure which shows the example of the measurement conditions (phase angle) of the ring test used for the setting, design, and manufacture of the motor drive system based on the same embodiment. FIG. 3 is a diagram showing an example of measurement results by a ring test used for setting, designing, and manufacturing the motor drive system according to the embodiment, in which (a) is a diagram showing fundamental wave current, and (b) is a diagram showing iron loss. It is. It is a figure which shows the rough flow of setting superimposition ratio n/m in the ring test used for the setting, design, and manufacturing of the motor drive system based on the same embodiment. It is a figure which shows the rough flow of setting initial phase (phi) of a 5th harmonic in the ring test used for the setting, design, and manufacturing of the motor drive system based on the same embodiment. FIG. 2 is a schematic configuration diagram of a motor testing device used for setting, designing, and manufacturing a motor drive system according to the same embodiment. FIG. 2 is a diagram showing an example of a test motor for a motor test used for setting, designing, and manufacturing a motor drive system according to the same embodiment, in which (a) is a schematic cross-sectional view of the motor, and (b) is a diagram showing specifications of the motor. It is. It is a figure which shows the example of the measurement conditions (superimposition rate) of the motor test used for the setting, design, and manufacture of the motor drive system based on the same embodiment. FIG. 3 is a diagram showing an example of a fundamental wave current and a fundamental sine wave modulation rate from a motor test used in setting, designing, and manufacturing the motor drive system according to the embodiment. It is a figure which shows the example of the total loss by the motor test used for the setting, design, and manufacture of the motor drive system based on the same embodiment. It is a figure which shows the example of the measurement conditions (phase angle) of the motor test used for the setting, design, and manufacturing of the motor drive system based on the same embodiment. FIG. 3 is a diagram showing an example of a fundamental sine wave modulation factor and fundamental wave current obtained by a motor test used in setting, designing, and manufacturing the motor drive system according to the embodiment. It is a figure which shows the example of the total loss by the motor test used for the setting, design, and manufacture of the motor drive system based on the same embodiment. It is a figure which shows the rough flow of setting superimposition ratio n/m in the motor test used for the setting, design, and manufacturing of the motor drive system based on the same embodiment. It is a figure which shows the rough flow of setting initial phase (phi) of the 5th harmonic in the motor test used for the setting, design, and manufacture of the motor drive system based on the same embodiment. FIG. 1 is a schematic configuration diagram of a test device related to characteristic evaluation test 1 (ring test) conducted to determine the configuration of the present invention. FIG. 3 is a diagram showing the specifications of a ring sample of the test device related to the characteristic evaluation test 1. It is a figure which shows the measurement conditions (superimposition rate characteristic) of the test device based on the same characteristic evaluation test 1. 2 is a diagram showing the waveform of a signal wave of the test device according to the characteristic evaluation test 1, in which (a) is a diagram showing a fundamental sine wave, and (b) is a diagram showing a fifth harmonic obtained by superimposing a fifth harmonic on the fundamental sine wave. FIG. 3 is a diagram showing a superimposed signal. It is a figure which shows the measurement result of the time waveform of the magnetic field strength and magnetic flux density by the test device based on the same characteristic evaluation test 1, (a) is a figure which shows the measured waveform when superimposition ratio n/m is -0.2. , (b) is a diagram showing the major loop component when the superposition ratio n/m is -0.2, (c) is a diagram showing the measured waveform when the superposition ratio n/m is -0.1, (d ) is a diagram showing the major loop component when the superimposition ratio n/m is −0.1. 3 is a diagram showing the measurement results of the time waveforms of the magnetic field strength and magnetic flux density by the test device according to the characteristic evaluation test 1, in which (a) shows the measured waveform when the superimposition ratio n/m is 0, and (b) ) is a diagram showing the major loop component when the superimposition ratio n/m is 0, (c) is a diagram showing the measured waveform when the superimposition ratio n/m is 0.2, (d) is a diagram showing the measurement waveform when the superimposition ratio n/m is 0.2. FIG. 3 is a diagram showing a major loop component when is 0.2. It is a figure showing the measurement results of the BH curve of magnetic field strength and magnetic flux density by the test device related to the same characteristic evaluation test 1, and (a) is the measurement result when the superimposition ratio n/m is -0.2 and the measure loop. A diagram showing the components, (b) is a diagram showing the measurement results and the major loop component when the superimposition ratio n/m is -0.1, (c) is a diagram showing the measurement results and the major loop component when the superimposition ratio n/m is 0. A diagram showing loop components, (d) is a diagram showing measurement results and major loop components when the superimposition ratio n/m is 0.2. FIG. 3 is a diagram showing measurement results of fundamental wave current and fundamental sine wave modulation factor by the test device according to the characteristic evaluation test 1. FIG. 2 is a diagram showing the measurement results of iron loss, major loop iron loss, and carrier harmonic iron loss (minor loop iron loss) by the test equipment related to Characteristic Evaluation Test 1, in which (a) is a diagram showing iron loss, and (b) is a diagram showing iron loss. ) is a diagram showing the rate of change in iron loss when the superimposition ratio n/m is 0 as a reference. FIG. 3 is a diagram showing measurement conditions (phase angle characteristics of fifth harmonic) of the test device according to the characteristic evaluation test 1. 2 is a diagram showing the waveform of a signal wave of the test device according to the same characteristic evaluation test 1, in which (a) is a diagram showing a fundamental sine wave, and (b) is a diagram showing a 5th order of the fundamental sine wave with an initial phase of π/4 [rad]. FIG. 3 is a diagram showing a fifth-order harmonic superimposed signal on which harmonics are superimposed. It is a figure which shows the measurement result when changing the initial phase of the 5th harmonic by the test device concerning the same characteristic evaluation test 1, (a) is a figure which shows the fundamental wave current, (b) is a figure which shows iron loss. It is a diagram. It is a figure which shows the measurement conditions (carrier frequency characteristic) of the test device based on the same characteristic evaluation test 1, (a) is a figure which shows the basic measurement condition, (b) is a figure which shows the superimposition condition of the 5th harmonic. FIG. 2 is a diagram showing the measurement results when the carrier frequency is changed by the test device according to the characteristic evaluation test 1, in which (a) is a diagram showing the fundamental wave current, and (b) is a diagram showing the iron loss. FIG. 2 is a schematic configuration diagram of a test device related to characteristic evaluation test 2 (motor test) conducted to determine the configuration of the present invention. It is a figure which shows the test motor of the test apparatus based on the same characteristic evaluation test 2, (a) is a schematic sectional drawing of a motor, (b) is a figure which shows the specification of a motor. It is a figure which shows the measurement conditions (superimposition rate characteristic) of the test device based on the same characteristic evaluation test 2. 2 is a diagram showing waveforms of three-phase signal waves of the test equipment related to the characteristic evaluation test 2, in which (a) is a diagram showing a fundamental sine wave, and (b) is a diagram showing a 5th harmonic wave superimposed on the fundamental sine wave. FIG. 3 is a diagram showing a harmonic superimposed signal. FIG. 6 is a diagram showing measurement results of fundamental wave current and fundamental sine wave modulation factor by the test device according to the characteristic evaluation test 2. FIG. 7 is a diagram showing the measurement results of the overall loss by the test device according to the characteristic evaluation test 2. FIG. 6 is a diagram showing the measurement results of motor core loss and mechanical loss by the test device according to the characteristic evaluation test 2. It is a figure which shows the measurement result of the copper loss and fundamental wave current copper loss by the test device based on the same characteristic evaluation test 2. FIG. 7 is a diagram showing the measurement results of inverter loss by the test device according to the characteristic evaluation test 2. FIG. 7 is a diagram showing the measurement conditions (phase angle characteristics of fifth harmonic) of the test device related to the characteristic evaluation test 2. 3 is a diagram showing waveforms of three-phase signal waves of the test equipment related to the same characteristic evaluation test 2, (a) is a diagram showing a fundamental sine wave, (b) is a diagram showing the fundamental sine wave with an initial phase of π/4 [rad] It is a figure which shows the 5th harmonic superimposition signal which superimposed the 5th harmonic of. FIG. 6 is a diagram showing the measurement results of the fundamental sine wave modulation factor and the fundamental wave current by the test device according to the characteristic evaluation test 2. FIG. 7 is a diagram showing the measurement results of the overall loss by the test device according to the characteristic evaluation test 2. FIG. 6 is a diagram showing the measurement results of motor core loss and mechanical loss by the test device according to the characteristic evaluation test 2. It is a figure which shows the measurement result of the copper loss and fundamental wave current copper loss by the test device based on the same characteristic evaluation test 2. FIG. 7 is a diagram showing the measurement results of inverter loss by the test device according to the characteristic evaluation test 2.

A motor drive system 1 according to an embodiment of the present invention will be explained using FIGS. 1 to 26. Further, the results of characteristic evaluation tests conducted to determine the configuration of the present invention will be explained using FIGS. 27 to 56.

FIG. 1 is a schematic configuration diagram of a motor drive system 1 according to an embodiment of the present invention. This motor drive system 1 is a system that drives a synchronous motor using a pulse width modulation (PWM, hereinafter referred to as PWM) control method, and includes a three-phase inverter section 2 (three-phase inverter circuit), It is configured to include a boost chopper section 3 (boost chopper circuit), a motor control section 4, a permanent magnet synchronous motor 5, a current sensor 9, and a position sensor 10.

The three-phase inverter section 2 switches the DC voltage supplied from the step-up chopper section 3 and performs pulse width modulation drive to become the motor drive voltage of three phases (U phase, V phase, W phase) of the permanent magnet synchronous motor 5. A voltage (hereinafter referred to as PWM drive voltage) is output. A three-phase PWM drive voltage output from the three-phase inverter section 2 is supplied to a three-phase stator coil 6 of the motor 5, and a permanent magnet rotor 7 is rotationally driven.

The three-phase inverter unit 2 includes switching elements S ₁ , S ₂ , S ₃ , S ₄ , S ₅ , and S ₆ made up of IGBTs (Insulated Gate Bipolar Transistors), and free-wheeling diodes D ₁ , D ₂ , D ₃ , There are three sets of configurations in which D ₄ , D ₅ , and D ₆ are connected in series as pairs on the upper and lower sides, and the upper and lower switching elements (S ₁ and S ₂ , S ₃ and S ₄ , Power for one phase of the motor ₅ is output from the connection point of the pair of freewheeling diodes (D ₁ and D ₂ , _D ₃ and D ₄ , D ₅ and D ₆ ). The upper and lower switching elements (S ₁ and S ₂ , S ₃ and S ₄ , S ₅ and S ₆ ) are driven so that if one is on, the other is off. Furthermore, in order to prevent short circuits, the upper and lower switching elements are driven to include a dead time in which both the upper and lower switching elements are turned off when switching between on and off, so that they are not turned on at the same time.

The on/off drive of the switching elements S ₁ to _S ₆ is performed by a motor control unit at the switching element input parts _{8 1} _, 8 ₂ , 8 ₃ , 8 ₄ , 8 ₅ , 8 _{6 which are the gate terminals of the switching elements S 1 to S 6} . This is performed by inputting the PWM drive signal supplied from 4.

Note that, in addition to IGBTs, MOSFETs (Metal-Oxide-Semiconductor Field Effect Transistors), power transistors, and the like can be used for the switching elements S ₁ to S ₆ .

The boost chopper section 3 supplies DC voltage to the three-phase inverter section 2. This boost chopper section 3 is composed of a battery 11, an inductor 12, a capacitor 13, a switching element Sc for the chopper section consisting of an IGBT, a diode 15, and the like. When the boost chopper control signal supplied from the motor control section 4 is input to the gate terminal of the switching element Sc for the chopper section, and this switching element Sc is turned on and off, due to the actions of the inductor 12, capacitor 13, and diode 15, , a DC voltage higher than the voltage of the battery 11 is output between both terminals of the capacitor 13. This output voltage is input to the collector terminals of the upper switching elements S ₁ , S ₃ , S ₅ of the three-phase inverter section 2, and becomes the input DC voltage to the three-phase inverter section 2. Depending on the level of the DC voltage output from the step-up chopper unit 3, the input DC voltage to the three-phase inverter unit 2 changes, and thereby the magnitude of the motor drive current flowing through the stator coil 6 of the permanent magnet synchronous motor 5 also changes. do.

Note that the step-up chopper section 3 is not essential, and the voltage of the battery 11, which is a DC power source, is directly supplied as the input DC voltage to the three-phase inverter section 2 without providing the step-up chopper section 3. The motor drive current to the permanent magnet synchronous motor 5 may be controlled by changing the modulation rate m, which is the amplitude of .

The current sensor 9 detects the motor drive current flowing through the stator coil 6 of the permanent magnet synchronous motor 5. The current value detected by this sensor 9 is output to the motor control section 4 and used to control the motor 5. As the current sensor 9, a shunt resistor and amplifier type sensor, a current sensor with a core, or a coreless current sensor can be used.

The permanent magnet synchronous motor 5 uses an interior permanent magnet synchronous motor (IPMSM). The motor 5 includes a stator having a stator coil 6 wound around a stator core made of a magnetic material, and a rotor 7 made of a permanent magnet rotatably supported inside the stator. The rotor 7 is rotationally driven by passing three-phase motor drive current through the stator coil 6 to form a rotating magnetic field. When the three-phase PWM drive voltage output from the three-phase inverter section 2 is applied to the three-phase stator coil 6, a motor drive current flows through the stator coil 6, causing the rotor 7 to rotate.

As the permanent magnet synchronous motor 5, a surface permanent magnet synchronous motor (SPMSM) can also be used. Also, other synchronous motors than this motor can be used.

The position sensor 10 detects the position of the magnetic poles of the permanent magnets constituting the rotor 7, and sensor detection signals corresponding to the U phase, V phase, and W phase are output to the motor control unit 4 and used to control the motor 5. used. A Hall IC or a Hall element can be used for the position sensor 10.

The motor control unit 4 receives the detection signals from the current sensor 9 and the position sensor 10, obtains the driving state of the permanent magnet synchronous motor 5, and compares it with command values such as a rotation speed command and a torque command, which are control commands for the motor 5. Then, the three-phase inverter section 2 and boost chopper section 3 are controlled so that the motor 5 operates in accordance with the command value. The three-phase inverter section 2 is controlled by outputting three-phase PWM drive signals to the switching element input sections 8 ₁ to 8 ₆ . This PWM drive signal is generated by comparing a signal wave (modulated wave) and a carrier wave (carrier wave) and switching at the intersection of the signal wave and the carrier wave.

Next, the motor control section 4 will be explained in detail.

FIG. 2 is a schematic block diagram of the motor control section 4 of the motor drive system 1. This motor control section 4 includes a CPU 40, a ROM 41, a RAM 42, a signal wave generation section 43, a carrier wave generation section 44, a PWM drive signal generation section 45, a PWM drive signal output section 46, a boost chopper control signal output section 47, a rotor detection position The configuration includes a receiving section 48, a motor input current value receiving section 49, and a command value receiving section 50.

The CPU 40 executes programs that control the motor drive system 1 and performs arithmetic processing. The ROM 41, which is a non-volatile memory, stores programs executed by the CPU 40 and data used for processing the programs. The RAM 42, which is a volatile memory, operates as a work area for program execution and arithmetic processing by the CPU 40.

The command value receiving unit 50 receives command values such as a rotational speed command and a torque command, which are control commands for the permanent magnet synchronous motor 5, from the outside. The motor control section 4 controls the three-phase inverter section 2 and the boost chopper section 3 to drive the motor 5 so as to match the received command value.

The motor input current value receiving unit 49 receives the detected value of the three-phase motor drive current output from the current sensor 9. Further, the rotor detection position receiving unit 48 receives a detected value of the magnetic pole position of the rotor 7 output from the position sensor 10. By detecting the position of the rotor 7 in time series, the rotational speed, rotational phase, etc. of the rotor 7 are calculated. The motor control unit 4 obtains the driving state of the permanent magnet synchronous motor 5 from the received motor drive current and the magnetic pole position of the rotor 7 .

In order to set the input DC voltage to the three-phase inverter section 2, the step-up chopper control signal output section 47 supplies a step-up chopper control signal to the switching element Sc for the chopper section of the step-up chopper section 3, and switches this switching element Sc. Drive on/off.

The signal wave generation section 43 generates a signal wave used as a base waveform when forming the PWM drive signals input to the switching element input sections 8 ₁ to 8 ₆ . The driving state of the permanent magnet synchronous motor 5 obtained from the signals of the current sensor 9 and the position sensor 10 is compared with the command value received by the command value reception unit 50, so that the operation of the motor 5 follows the command value. The CPU 40 performs calculations to set the modulation rate and phase, which is the amplitude of the signal wave. This signal wave is generated as a three-phase waveform corresponding to the U phase, V phase, and W phase. The signal wave is generated as numerical data by the calculation of the CPU 40.

The carrier wave generation unit 44 generates a carrier wave used as a base waveform when forming a PWM drive signal. This carrier wave is generated as numerical data by the calculation of the CPU 40.

The PWM drive signal generation unit 45 generates a PWM drive signal such that the edge position is switched at the intersection of the signal wave generated by the signal wave generation unit 43 and the carrier wave generated by the carrier wave generation unit 44. This PWM drive signal is generated as a three-phase signal corresponding to the U phase, V phase, and W phase. The PWM drive signal is generated by the calculation of the CPU 40.

The PWM drive signal output section 46 supplies the three-phase PWM drive signal generated by the PWM drive signal generation section 45 to the switching element input sections 8 ₁ to 8 ₆ . By inputting this PWM drive signal, a three-phase PWM drive voltage is output from the three-phase inverter section 2 to the permanent magnet synchronous motor 5.

The PWM drive signal supplied to the upper and lower switching elements (S ₁ and S ₂ , S ₃ and S ₄ , S ₅ and S ₆ ) is applied to the upper and lower switching elements (S ₁ and S ₂ , S ₃ and S ₄ , S ₅ and S ₆ ) is turned on, the other is driven to be turned off, so that the waveforms are reversed on the upper and lower sides. However, a dead time is provided to prevent the upper and lower switching elements from being turned on at the same time.

The generation of the signal wave, carrier wave, and PWM drive signal is performed by the CPU 40 based on a program stored in the ROM 41 and calculated numerically. With this software configuration, parameters for forming signal waves and carrier waves can be changed flexibly and easily. However, the generation of the signal wave, carrier wave, and PWM drive signal is not limited to the software configuration, and the generation of the signal wave, carrier wave, and PWM drive signal may be realized using an electronic circuit as a hardware configuration.

FIG. 3 is a diagram illustrating pulse width modulation of the PWM drive signal of the motor drive system 1, in which (a) is a diagram showing a signal wave and a carrier wave, (b) is a diagram showing a PWM drive signal, and (c) is a diagram showing a PWM drive signal. is a diagram showing PWM drive voltage.

FIG. 3A shows a signal wave h(t) generated by the signal wave generation unit 43 and a carrier wave c(t) generated by the carrier wave generation unit 44.

The signal wave h(t) is generated by adding and superimposing the fifth harmonic to the fundamental sine wave g(t). If the frequency of this fundamental sine wave g(t) is the fundamental sine wave frequency _f1 , the frequency of the fifth harmonic is five times 5· _f1 . Furthermore, when the modulation rate expressed as the amplitude of the fundamental sine wave g(t) is m, and the modulation rate expressed as the amplitude of the fifth harmonic is n, 5 for the modulation rate m of the fundamental sine wave g(t). The ratio of the modulation rate n of the harmonics is defined as the superimposition rate n/m.

The carrier wave c(t) is a triangular wave with a carrier frequency _fc , and the amplitude of the triangular wave is 1. The modulation rate m of the fundamental sine wave g(t) takes a smaller value than the amplitude of the carrier wave c(t).

In this figure, the horizontal axis shows the phase converted from time t, and the vertical axis shows the magnitude of the signal.

The PWM drive signal generation unit 45 performs switching at the intersection of the signal wave h(t) and the carrier wave c(t), modulates the pulse width, and generates a PWM drive signal. That is, a PWM drive signal is generated by switching the pulse width at the intersection of the signal wave h(t) and the carrier wave c(t).

FIG. 3(b) shows the PWM drive signal generated by the PWM drive signal generation section 45. In this figure, the horizontal axis represents the phase converted from time t, and the vertical axis represents the magnitude of the PWM drive signal. The PWM drive signal is composed of a rectangular waveform, and in comparing the signal wave h(t) and the carrier wave c(t) in FIG. ), it becomes a high level, and when the magnitude of the signal wave h(t) is smaller than the carrier wave c(t), it becomes a low level. A rising edge or a falling edge is formed at the intersection of the signal wave h(t) and the carrier wave c(t).

When the magnitude of the signal wave h(t) increases within one period of the signal wave h(t), the pulse width indicated as the high level width increases, and the magnitude of the signal wave h(t) increases. When it becomes smaller, the pulse width, which is the high level width, becomes narrower. In this manner, the time period during which the pulse width of the PWM drive signal is at a high level changes periodically within one period of the signal wave h(t).

Regarding periodic fluctuations in the pulse width of the PWM drive signal, the fundamental sine wave frequency _f1 included in the signal wave h(t) becomes the fundamental frequency of the fluctuation in the pulse width of the PWM drive signal, and the signal wave h(t) The frequency component of the fifth harmonic included in is added.

When the PWM drive signal shown in FIG. 3(b) is output from the PWM drive signal output section 46 and supplied to the switching element input sections 8 ₁ to 8 ₆ of the three-phase inverter section 2, the upper and lower switching A permanent magnet is connected to the connection point of a pair of elements (S ₁ and S ₂ , S ₃ and S ₄ , S ₅ and S ₆ ) and free-wheeling diodes (D ₁ and D ₂ , D ₃ and D ₄ , D ₅ and D ₆ ). A PWM drive voltage is output as the drive voltage of the synchronous motor 5.

In this way, the switching operation of the semiconductor in the three-phase inverter unit 2 is performed at the intersection of the signal wave h(t) and the carrier wave c(t), and a PWM drive voltage is output as the drive voltage of the permanent magnet synchronous motor 5. let

FIG. 3(c) shows the PWM drive voltage output from the three-phase inverter section 2. In this figure, the horizontal axis represents the phase converted from time t, and the vertical axis represents the magnitude of the PWM drive voltage. The waveform of the PWM drive voltage has a similar shape to the waveform of the PWM drive signal on the time (phase) axis.

In this way, the pulse width indicated as the high-level width of the PWM drive voltage also repeats periodic fluctuations, similar to the PWM drive signal. Regarding periodic fluctuations in the pulse width of the PWM drive voltage, the fundamental sine wave frequency _f1 included in the signal wave h(t) becomes the fundamental frequency of the fluctuation in the pulse width of the PWM drive voltage, and this is in addition to the signal wave h(t ) is added with the frequency component of the fifth harmonic included in .

The fundamental sine wave frequency _f1 included in the signal wave h(t) defines the fundamental frequency of the pulse width of the PWM drive voltage. In addition, by using the signal wave h(t) on which the fifth harmonic is superimposed, the intersection point with the carrier wave c(t) changes depending on the fifth harmonic component, and the PWM drive output from the three-phase inverter section 2 A fifth harmonic is superimposed on the voltage.

The PWM drive voltage is output from the three-phase inverter section 2 as a three-phase waveform corresponding to the U phase, V phase, and W phase of the permanent magnet synchronous motor 5, but since these three phases are configured, the initial phases are different from each other. A three-phase PWM drive signal is generated using three signal waves shifted by 2π/3 [rad].

FIG. 4 shows fundamental sine waves g _u (t), g _v (t), and g _w (t) included in the three-phase signal waves. In this figure, the horizontal axis shows the phase converted from time t, and the vertical axis shows the magnitude of the signal. To generate _the PWM drive signal, a signal wave h _u ₍ _t ), h _v (t), h _w (t), and as mentioned above, these signal waves h _u (t), h _v (t), h _w (t) and the triangular carrier wave c (t) It is formed so that the pulse width is switched at the intersection with

These three-phase fundamental sine waves g _u (t), g _v (t), g _w (t) and three-phase signal waves h _u (t), h _v (t), h _w (t) are as follows. Expressions (1) to (6) are shown below.
g _u (t) is a U-phase fundamental sine wave among three-phase fundamental sine waves, g _v (t) is a V-phase fundamental sine wave, and g _w (t) is a W-phase fundamental sine wave.

h _u (t) is a U-phase signal wave among three-phase signal waves, h _v (t) is a V-phase signal wave, and h _w (t) is a W-phase signal wave.

Here, f ₁ is the fundamental sine wave frequency of the fundamental sine waves g _u (t), g _v (t), g _w (t), and m is the fundamental sine wave frequency g _u (t), g _v (t ), the modulation factor expressed as the amplitude of g _w (t). Further, n is a modulation rate expressed as the amplitude of the fifth harmonic to be superimposed, and φ is the initial phase of this fifth harmonic. And t is time.

As will be described later, the upper limit of the superposition ratio n/m is set to the value n/m at which the fundamental wave currents I _f1 and I _{f1_rms} start to decrease due to changes in magnetic properties, and the lower limit of the superposition ratio n/m is The value n/m is set such that the increase due to the harmonic component is greater than the reduction effect of the fundamental wave currents I _f1 and I _{f1_rms} , and the superimposition ratio n/m operates within this lower limit and upper limit.

It is preferable to operate by setting the superimposition ratio n/m to a range of −0.3 or more and less than 0.

Furthermore, when the phase angle of the fundamental sine wave g(t) shown in FIG. 3(a) is π/2 radian (rad), the numerical value of the fundamental sine wave g(t) is the signal wave h(t) It is preferable to operate by setting the initial phase φ of the fifth harmonic to be equal to or greater than the value of .

For example, when explaining _using the fundamental sine wave g _u (t) shown in equation (1) and the signal wave h _u (t) shown in equation (2), the phase angle (2πf ₁ The numerical value m・sin(π/2) of the fundamental sine wave g _u (t) when t) is π/2 radian is the signal wave h _u (t)=m・sin(π/2)+n・sin( The initial phase φ is set to be equal to or greater than the numerical value of 5·π/2+φ). In other words, the condition that g _u (t) (= m・sin (π/2)) ＞= h _u (t) (= m・sin (π/2) + n・sin (5・π/2 + φ)) is The initial phase φ of the fifth-order harmonic is set so as to satisfy the following conditions.

Further, the maximum phase angle of the initial phase φ is set to the value of the initial phase φ at which the fundamental wave currents I _f1 and I _{f1_rms} begin to decrease due to changes in magnetic properties, and the minimum phase angle of the initial phase φ The initial phase φ may be set to a value where the increase due to the harmonic component is greater than the reduction effect of the currents I _f1 and I _{f1_rms} , and the operation may be performed within the range between the minimum phase angle and the maximum phase angle.

In the characteristic evaluation test described later, the basic sine waves g _u (t), g _v (t), g _w (t) are used as signal waves, and the basic sine waves g _u (t), g _v (t) are used as signal waves. , g _w (t) with a fifth-order harmonic superimposed thereon, the losses are compared when signal waves h _u (t), h _v (t), and h _w (t) are used.

As a result of this characteristic evaluation test, the superposition ratio n/m is -0.25 or more and -0.05 or less, and the initial phase φ of the fifth harmonic is -π/4 [rad] or more and π/2 [rad] Under the following conditions, the loss is reduced compared to when the fundamental sine waves g _u (t), g _v (t), and g _w (t) are used as signal waves, and setting this condition , it was shown that it is preferable to operate it.

Also, based on this condition, the possible range of the superimposition ratio n/m is set to -0.15 or more and -0.1 or less, and the possible range of the initial phase φ is set to π/8 [rad] or more and 5π It was shown that setting the value to /16 [rad] or less further reduces the loss and is more preferable.

Furthermore, if the superimposition ratio n/m is set in the range of -0.15 or more and -0.1 or less, and the initial phase φ of the fifth harmonic is set to π/4 [rad], the loss reduction effect is was shown to be even higher and more preferable.

Furthermore, the harmonics to be superimposed on the fundamental sine waves g _u (t), g _v (t), and g _w (t) can be of the 5th or higher order, and the a-th harmonic that is a times the fundamental sine wave frequency f ₁ Three-phase signal waves h _ua (t), h _va (t), and h _wa (t) in the case of superimposing waves are shown as in equations (7) to (9).
Here, n _a is the modulation rate expressed as the amplitude of the fifth harmonic to be superimposed, and φ _a is the initial phase of this a harmonic.

The upper limit of the superposition ratio n _a /m is set to the value of n _a /m at which the fundamental wave currents I _f1 and I _{f1_rms} start to decrease due to changes in magnetic properties, and the lower limit of the superposition ratio n _a /m is set to the value The wave current I _f1 , I _{f1_rms} is set to a value of n _a /m where the increase due to the harmonic component is greater than the reduction effect, and the superimposition rate n _a /m operates within this lower limit and upper limit.

It is preferable to operate by setting the superimposition ratio n _a /m to a range of -0.3 or more and less than 0.

Also, when the phase angle of the fundamental sine wave g(t) shown in FIG. 3(a) is π/2 radian, the value of the fundamental sine wave g(t) is the value of the signal wave h _a (t) at that time. It is preferable to operate by setting the initial phase φ _a of the a-th harmonic as above.

For example, when explaining using the fundamental sine wave g _u (t) shown in equation (1) and the signal wave h _ua (t) shown in equation ( ₇ ), the phase angle (2πf ₁ The numerical value m・sin(π/2) of the fundamental sine wave g _u (t) when t) is π/2 radian is the signal wave h _ua (t)=m・sin(π/2)+n _a・sin It operates by setting the initial phase φ _a to be equal to or greater than the value of (a·π/2+φ _a ). In other words, the condition that g _u (t)(=m・sin(π/2)) ＞= h _ua (t)(=m・sin(π/2)+n・sin(a・π/2+φ)) is The initial phase φ _a of the a-th harmonic is set so as to satisfy the following conditions.

Further, the maximum phase angle of the initial phase φ _a of the a-th harmonic is set to the value of the initial phase φ _a where the fundamental wave current I _f1 , I _{f1_rms} starts to decrease due to a change in the magnetic characteristics, and the initial phase φ _a The minimum phase angle of is set to the value of the initial phase φ _a at which the increase due to the harmonic component is greater than the reduction effect of the fundamental wave current I _f1 , I _{f1_rms} , and the operation is performed within the range of this minimum phase angle and the maximum phase angle. You can do it like this.

By analogy with the results of the characteristic evaluation test described later using signal waves h _u (t), h _v (t), and h _w (t) on which fifth-order harmonics are superimposed, the superposition ratio n _a /m is -0. .25 or more and -0.05 or less, and when the initial phase φ _a of the a-th harmonic is -π/4 [rad] or more and π/2 [rad] or less, the fundamental sine wave g _u ( t), g _v (t), and g _w (t) as signal waves, it is estimated that the loss will be reduced, so it is preferable to set and operate under these conditions.

Also, based on this condition, the possible range of the superposition ratio n _a /m is set to -0.15 or more and -0.1 or less, and the possible range of the initial phase φ _a is set to π/8 [rad] or more. In addition, it is preferable to set it to 5π/16 [rad] or less because it is estimated that the loss will be further reduced by analogy with the results of the characteristic evaluation test.

In addition, when operating with the superimposition ratio n _a /m set to a range of -0.15 or more and -0.1 or less, and the initial phase φ _a of the a-th harmonic set to π/4 [rad], By analogy with the results of the characteristic evaluation test, it can be inferred that the loss reduction effect will be even higher, which is preferable.

In addition, in the case of a three-phase motor, it is thought that there will be no effect even if even-numbered harmonics and multiples of 3 are superimposed on the a-th harmonics of the fifth or higher harmonics, so other harmonics, such as the seventh It is effective to superimpose harmonics such as , 11th, and 13th harmonics.

Next, the operation of the motor drive system 1 according to the embodiment of the present invention will be explained.

FIG. 5 is a diagram showing the flow of operation of the motor control section 4 of this motor drive system 1.

The motor control unit 4 receives command values such as a rotation speed command and a torque command, which are control commands for the permanent magnet synchronous motor 5, from the outside in a command value reception unit 50 (step S1), and stores the received command values in the RAM 42. (S2 step). Thereafter, when a motor rotation start instruction is input from the outside, the motor control unit 4 receives this rotation start instruction (step S3) and starts the operation of driving the motor 5 to rotate.

The motor control unit 4 receives the magnetic pole position of the rotor 7 output from the position sensor 10 and the detected value of the three-phase motor drive current output from the current sensor 9 (step S4). Based on the magnetic pole position of the rotor 7 and the detected value of the motor drive current, the driving state of the permanent magnet synchronous motor 5 is acquired. The motor control unit 4 performs control so that the drive of the motor 5 matches the command value.

First, a step-up chopper control signal is generated so that the input DC voltage to the three-phase inverter section 2 becomes an appropriate value (step S5), and the generated step-up chopper control signal is applied to the chopper section switching of the step-up chopper section 3. Output to element Sc (step S6).

Next, the signal waves h _u (t), h _v (t), h _w ( t) (signal wave generation processing) (step S7). Specifically, by setting the fundamental sine wave frequency f ₁ , phase, and modulation rate m of the fundamental sine waves g _u (t), g _v (t), and g _w (t), this fundamental sine wave g _u ( t), g _v (t), and g _w (t) with a modulation rate n and a fifth harmonic of the initial phase φ is superimposed on them.

Next, a carrier wave configured as a triangular wave is generated (carrier wave generation process) (step S8).

Thereafter, as shown in FIG. 3, a PWM drive signal is generated so that the pulse width is switched at the intersection of the signal waves h _u (t), h _v (t), h _w (t) and the carrier wave (PWM drive signal processing) (S9 step).

Then, the generated PWM drive signal is output to the switching element input sections 8 ₁ to 8 ₆ of the three-phase inverter section 2 (PWM drive supply process) (step S10). As a result, the switching elements (S ₁ and S ₂ , S ₃ and S ₄ , S ₅ and S ₆ ) on the upper and lower sides of the three-phase inverter section 2 and the freewheeling diodes (D ₁ and D ₂ , D ₃ and D ₄ , D ₅ and D ₆ ), a PWM drive voltage is output as the drive voltage of the permanent magnet synchronous motor 5.

The motor control unit 4 receives a new command value from the outside and determines whether there is a change in the command value with respect to the previously stored command value (step S11). If there is a change in the command value (Yes in step S11), the changed command value is stored in the RAM and updated (step S12). If there is no change in the command value (No in step S11), the operation continues without doing anything.

Thereafter, the steps S4 to S12 are repeated to drive the permanent magnet synchronous motor 5 so as to match the command value.

Next, the setting method, design method, and manufacturing method of the motor drive system 1 according to the embodiment of the present invention will be explained.

In this setting method, design method, and manufacturing method, the superposition ratio of the signal wave h(t) shown in equation (11) obtained by superimposing the fifth harmonic on the fundamental sine wave g(t) shown in equation (10) is The upper and lower limits of n/m, the set value of the superimposition ratio n/m, the maximum and minimum phase angles of the initial phase φ of the fifth harmonic, and the set value of the initial phase φ are determined.
Regarding the setting method, design method, and manufacturing method of the motor drive system 1, first, a method using a ring test will be explained, and then a method using a motor test will be explained. In both the ring test method and the motor test method, measurements are performed while changing the superimposition ratio n/m.

<Setting method, design method, and manufacturing method using ring test>
First, a method of setting, designing, and manufacturing the motor drive system 1 using a ring test will be explained.

FIG. 6 is a schematic configuration diagram of a ring testing device 109 used for setting, designing, and manufacturing this motor drive system 1. This ring test device 109 has the same configuration as the ring test device 69 used in characteristic evaluation test 1 (ring test) described later (therefore, refer to characteristic evaluation test 1 (ring test) described later for details).

FIG. 7 is a diagram showing an example of the setting conditions for the ring test, in which (a) is a diagram showing the specifications of the ring sample 101, and (b) is a diagram showing the measurement conditions (overlapping ratio). The same material as the iron core material (core material) of the rotor and stator of the permanent magnet synchronous motor 5 used in the motor drive system 1 is used for the iron core material (core material) that is the ring sample 101 for this ring test. That is, the ring sample 101 is made of the same material as the iron core material of the permanent magnet synchronous motor 5 used in the motor drive system 1.

The IGBT inverter 102 shown in FIG. 6 is a single-phase Si-IGBT inverter. This IGBT inverter 102 includes Si-IGBTs as switching elements S ₁₀₁ , S ₁₀₂ , S ₁₀₃ , and S ₁₀₄ and Si diodes as freewheeling diodes D ₁₀₁ , D ₁₀₂ , D ₁₀₃ , and D ₁₀₄ . The switching elements S ₁₀₁ to S ₁₀₄ and the free wheel diodes D ₁₀₁ to D ₁₀₄ used in the IGBT inverter 102 of the ring test device 69 are the same as the switching elements S 1 to S 104 used in the three-phase inverter section 2 of the motor drive system ₁ . S ₆ and the free wheel diodes D ₁ to D ₆ are the same.

The measurement conditions are, for example, a fundamental sine wave frequency f ₁ of 50 [Hz], a carrier frequency f _c of 1 [kHz], and an input voltage V _dc supplied from the DC power supply 103 to the IGBT inverter 102 of 15 [V]. Then, the fundamental sine wave modulation rate m is adjusted so that the fundamental sine wave magnetic flux density B _f1 becomes 1 [T]. It is considered that the constant fundamental sinusoidal magnetic flux density B _f1 in the ring test corresponds to the constant average torque in the motor test.

A fifth harmonic superimposition PWM method is adopted for controlling the IGBT inverter 102. That is, a PWM signal is generated by switching the pulse width at the intersection of the signal wave h(t) on which the fifth harmonic shown in the above equation (11) is superimposed and the carrier wave configured as a triangular wave. is input to the switching element input sections 108 ₁ and 108 ₂ which are the gate terminals of the pair of upper and lower switching elements (S ₁₀₁ and S ₁₀₂ ) of the IGBT inverter 102 . In addition, a PWM signal is generated by switching the pulse width at the intersection of the vertically inverted signal wave h(t) and the carrier wave, and this PWM signal is transmitted to the pair of upper and lower switching elements (S ₁₀₃ and S ₁₀₄ ) of the IGBT inverter 102. ) are input to switching element input sections 108 3 and 108 ₄ which are the gate terminals of the switching elements 108 ₃ and 108 4 . At this time, the upper and lower switching elements (upper and lower S ₁₀₁ and S ₁₀₂ , and upper and lower S ₁₀₃ and S ₁₀₄ ) are driven so that if one is on, the other is off. In addition, to prevent short circuits, to prevent the upper and lower switching elements (upper and lower S ₁₀₁ and S ₁₀₂ , and upper and lower S ₁₀₃ and S ₁₀₄ ) from turning on at the same time, there is a dead time in which both the top and bottom are turned off when switching between on and off. is driven to include.

In this way, the upper and lower switching elements ₍ _{S 101} _and S ₁₀₂ , S ₁₀₃ _and _S ₁₀₄ The pulse width modulated voltage input to the primary coil is output from the connection point of the pair of freewheeling diodes (D ₁₀₁ and D ₁₀₂ , D ₁₀₃ and D ₁₀₄ ). The application of this pulse width modulated voltage causes a primary current _I1 to flow in the primary coil.

Then, a primary current I ₁ flowing through the primary coil wound around the ring sample 101 and a secondary voltage V ₂ generated at the secondary coil wound around the ring sample 101 are measured.

Next, a method for calculating iron loss in a ring test will be explained. The primary current I ₁ and secondary voltage V ₂ shown in FIG. 6 are measured, and the magnetic field strength H and magnetic flux density B are determined as shown in equations (12) and (13).
Using this magnetic field strength H and magnetic flux density B, iron loss P _fe is determined as shown in equation (14).
Here, since the magnetic field strength H and magnetic flux density B determined by the above equations (12) and (13) include carrier harmonic components, the magnetization phenomenon becomes complicated. Therefore, in order to clarify the influence of fifth-order harmonic superposition, we investigated the low-order frequency components (f ₁ , f ₃ , f ₅ components: f ₃ is the frequency of the 3rd harmonic of the fundamental sine wave frequency f ₁ , and f ₅ is Consider extracting the fundamental sine wave frequency f (fifth harmonic frequency of ₁ ). The obtained magnetic field strength H and magnetic flux density B are fitted using cftool (approximation curve tool) by the numerical calculation software MATLAB (registered trademark) R2019b (The MathWorks, Inc.) to calculate the magnetic field strength. A major loop component H _major and a major loop component B _major of magnetic flux density are calculated. From these, the major loop iron loss P _major is calculated as shown in equation (15). Further, as in equation (16), the difference between the iron loss P _fe and the major loop iron loss P _major is defined as a carrier harmonic iron loss (minor loop iron loss) P _carrier .
Below, examples of measurement results in a ring test measured while changing the superimposition ratio n/m are shown.

FIG. 8 is a diagram showing an example of the time waveform of the magnetic field strength H and magnetic flux density B in the ring test, (a) is a diagram showing the measured waveform when the superimposition ratio n/m is 0, (b) is a diagram showing the major loop components (H _major , B _major ) when the superimposition ratio n/m is 0, (c) is a diagram showing the measured waveform when the superimposition ratio n/m is -0.2, (d ) is a diagram showing major loop components (H _major , B _major ) when the superimposition ratio n/m is −0.2.

When the superimposition ratio n/m is 0, this corresponds to the case where the fundamental sine wave g(t) is used as the signal wave h(t) without adding the fifth harmonic. Comparing the cases where the superposition ratio n/m is 0 and -0.2, the time waveform shapes of the magnetic field strength H and magnetic flux density B are different, and changes appear in the magnetization phenomenon due to the fifth harmonic superposition. I understand that.

FIG. 9 is a diagram showing an example of a BH curve of magnetic field strength H and magnetic flux density B in a ring test, where (a) shows the measurement results (solid line) when the superimposition ratio n/m is 0 and the major loop component. (dashed line), (b) is a diagram showing the measurement results (solid line) and the major loop component (dashed line) when the superimposition ratio n/m is -0.2, (c) is a diagram showing the measurement results when the superimposition ratio n/m is -0.2, and (c) is a diagram showing the measurement results when the superimposition ratio n/m is FIG. 3 is a diagram showing major loop components of 0 (broken line) and −0.2 (solid line).

This figure shows the measurement data shown in FIG. 8, with the horizontal axis representing the magnetic field strength H and the vertical axis representing the magnetic flux density B. As shown in FIG. 9(c), when comparing the cases where the superimposition ratio n/m is 0 and -0.2, the BH curve of the major loop component H _major of the magnetic field strength and the major loop component B _major of the magnetic flux density is The shapes of the two are different, and it can be seen that a change appears in the magnetization phenomenon due to fifth-order harmonic superposition. In other words, "changes in magnetic properties" occur due to fifth-order harmonic superposition. The BH curve of the major loop component shown in FIG. 9(c) shows the points where the fundamental sinusoidal magnetic flux density B _f1 becomes 1 [T] when the superimposition rate n/m is 0 and -0.2, respectively. The plots are marked with × and *. The major loop component changes greatly due to the superposition of the fifth harmonic, and accordingly, the position marked *, which is a condition where the superposition ratio n/m is -0.2, changes to a point where the magnetic permeability increases. That is, the fundamental sinusoidal magnetic flux density B _f1 reaches 1 [T] at the point where the magnetization phenomenon changes due to the fifth-order harmonic superposition and the magnetic permeability increases.

FIG. 10 is a diagram showing an example of the fundamental wave current I _f1 and the fundamental sine wave modulation factor m obtained by the ring test.

The fundamental wave current I _f1 in the ring test is the effective value of the component of the fundamental sine wave frequency f ₁ of the primary current I ₁ flowing through the primary coil.

This fundamental wave current I _f1 is an excitation current component for obtaining the fundamental wave magnetic flux density B _f1 , and increases slightly when the superposition rate n/m is greater than 0 compared to when the superposition rate n/m is 0. , tends to decrease significantly when the superimposition ratio n/m is smaller than 0.

When the superimposition ratio n/m becomes smaller than 0, "the fundamental wave current I _f1 starts to decrease due to a change in magnetic properties." The value of this superimposition ratio n/m may be set as an upper limit. Furthermore, in the range where the superposition ratio n/m is -0.1 or less, the fundamental wave current I _f1 decreases rapidly, and when this superposition ratio n/m is -0.1, it is defined as "change in magnetic properties." It can also be said to be the value of the superimposition ratio n/m at which the fundamental wave current I _f1 starts to decrease at . The value of this superimposition ratio n/m may be set as an upper limit.

Further, the basic sine wave modulation rate m is adjusted so that the basic sine wave magnetic flux density B _f1 becomes 1 [T] under a constant DC voltage V _dc to the IGBT inverter 102, and the superimposition rate n/m is 0. When the superimposition ratio n/m becomes larger, the fundamental sine wave modulation rate m increases, and when the superimposition ratio n/m becomes smaller than 0, the fundamental sine wave modulation rate m decreases.

It can be said that due to the superposition of the fifth harmonic, the fundamental wave current I _f1 decreases when the fundamental sinusoidal magnetic flux density B _f1 is constant.

FIG. 11 is a diagram showing an example of iron loss P _fe , major loop iron loss P _major , and carrier harmonic iron loss (minor loop iron loss) P _carrier in a ring test; (a) is a diagram showing iron loss; (b) is a diagram showing the rate of change in iron loss when the superimposition ratio n/m is 0 as a reference.

As shown in Fig. 11(a), compared to the case where the superposition ratio n/m is 0, the iron loss P _fe increases in the range where the superposition ratio n/m is greater than 0, and the superposition ratio n/m is smaller than 0. In this range, the iron loss P _fe decreases. Furthermore, the major loop iron loss P _major is reduced when the superimposition ratio n/m is -0.15, -0.1, and -0.05, compared to when the superposition ratio n/m is 0. Compared to the case where the superposition ratio n/m is 0, the carrier harmonic iron loss P _carrier increases when the superposition ratio n/m is greater than 0, and decreases when the superposition ratio n/m is smaller than 0.

FIG. 11(b) shows the results when the superposition ratio n/m is changed based on the iron loss P _fe , major loop iron loss P _major , and carrier harmonic iron loss P _carrier when the superposition ratio n/m is 0. It shows the rate of change of iron loss P _fe , major loop iron loss P _major , and carrier harmonic iron loss P _carrier .

In a range where the superimposition ratio n/m is larger than 0, the iron loss P _fe increases, and in a range where the superimposition ratio n/m is smaller than 0, the iron loss P _fe decreases. Further, when the superimposition ratio n/m was −0.2, the iron loss P _fe showed the minimum value, and the iron loss reduction rate was 3.3%. Further, the iron loss P _fe when the superimposition ratio n/m is -0.25 is slightly higher than when the superimposition ratio n/m is -0.2.

The major loop iron loss P _major decreases when the overlap ratio n/m is −0.15, −0.1, and −0.05. When the superimposition ratio n/m is −0.1, the major loop iron loss P _major is at its minimum, and the reduction rate is 2.3%. The carrier harmonic iron loss P _carrier increases when the superposition ratio n/m is greater than 0, and decreases when the superposition ratio n/m is smaller than 0. When the superimposition ratio n/m is −0.25, it is minimum, and the iron loss reduction rate of the carrier harmonic iron loss P _carrier is 17.8%.

In this way, as the superimposition ratio n/m is decreased from the larger side (for example, superimposition ratio n/m = 0.25), the iron loss P _fe decreases, and when the superimposition ratio n/m becomes smaller than 0, the superposition ratio Iron loss P _fe is reduced compared to when n/m is 0. Then, when the superimposition ratio n/m is further reduced, the iron loss P _fe becomes minimum when the superimposition ratio n/m is -0.2, and thereafter the iron loss P _fe tends to increase. When extrapolating the graph of the rate of change of iron loss P _fe shown in FIG. 11(b) to a range where the superimposition ratio n/m is smaller than -0.25, it is found that in a range where the superimposition ratio n/m is -0.3 or more, , it can be said that the iron loss P _fe is reduced compared to the case where the fifth harmonic is not superimposed (when the superimposition ratio n/m is 0). Then, compared to the case where the superimposition ratio n/m is 0, the range in which the iron loss P _fe is reduced is the range where the superimposition ratio n/m is −0.3 or more and less than 0. When the superimposition ratio n/m is -0.3, it can be said that the value of the superposition ratio n/m is such that "the increase due to the harmonic component is greater than the reduction effect of the fundamental wave current I _f1 ." This value may be set as the lower limit of the superimposition ratio n/m.

FIG. 12 is a diagram showing an example of iron loss P _fe and fundamental wave current I _f1 obtained by a ring test. This FIG. 12 summarizes measurement results different from those in FIGS. 10 and 11.

Compared to the case where the superimposition ratio n/m is 0, the iron loss P _fe increases in the range where the superimposition ratio n/m is larger than 0, and the iron loss P _fe decreases in the range where the superimposition ratio n/m is smaller than 0. ing. Further, when the superimposition ratio n/m is −0.15, the iron loss P _fe is the minimum.

As the superposition ratio n/m decreases from the larger side, the iron loss P _fe decreases, and when the superposition ratio n/m becomes smaller than 0, the iron loss P _fe decreases compared to when the superposition ratio n/m is 0. do. Then, when the superimposition ratio n/m is further reduced, the iron loss P _fe becomes minimum when the superimposition ratio n/m is -0.15, and thereafter the iron loss P _fe tends to increase. If we extrapolate the graph of iron loss P _fe shown in Figure 12 to the range where the superposition ratio n/m is smaller than -0.2, we can see that the fifth harmonic is It can be said that the iron loss P _fe is reduced compared to the case where there is no overlap (the case where the overlap ratio n/m is 0). Then, compared to the case where the superposition ratio n/m is 0, the range in which the iron loss P _fe is reduced is the range where the superposition ratio n/m is −0.3 or more and smaller than 0.

The fundamental wave current I _f1 tends to increase slightly when the superposition ratio n/m is greater than 0, and to decrease significantly when the superposition ratio n/m is smaller than 0, compared to when the superposition ratio n/m is 0. .

Below, examples of measurement results in a ring test measured while changing the initial phase φ of the fifth harmonic are shown.

FIG. 13 is a diagram showing an example of measurement conditions (phase angle) for a ring test in which measurement is performed by changing the initial phase φ of the fifth harmonic.

FIG. 14 is a diagram showing an example of measurement results by a ring test, in which (a) is a diagram showing the fundamental wave current I _f1 , and (b) is a diagram showing the iron loss P _fe . The horizontal broken line in the figure indicates the fundamental wave current _{I f1} and the iron loss when the modulation rate n of the fifth harmonic is 0, that is, when the fundamental sine wave g(t) is used as the signal wave h(t). Measured values of P _fe are shown.

Compared to the horizontal broken line, the fundamental wave current I _f1 decreases when the initial phase φ is 0 [rad] or less, and increases when the initial phase φ is π/4 [rad] or more.

When the initial phase φ becomes smaller than π/4 [rad], it can be said that "the fundamental wave current I _f1 starts to decrease due to a change in magnetic characteristics." The value of this initial phase φ (π/4) may be set as the maximum phase angle.

Compared to the horizontal broken line, the iron loss P _fe is reduced in the range where the initial phase φ is -π/4 [rad] or more and π/2 [rad] or less, and when the initial phase φ is π/4 [rad] ] is the minimum value.

In this way, by decreasing the initial phase φ of the fifth harmonic from the large side (for example, initial phase φ = 3π/4 [rad]), the iron loss P _fe decreases, and the initial phase φ becomes π/2 [rad]. ], the iron loss P _fe is lower than when the superimposition ratio n/m is 0. As the initial phase φ is further reduced, the iron loss P _fe becomes the minimum when the initial phase φ is π/4 [rad], and thereafter the iron loss P _fe increases. When the initial phase φ becomes -π/4 [rad], it shows almost the same value as the iron loss P _fe when the superposition ratio n/m is 0, and when the initial phase φ is further reduced, the superposition ratio n/m The iron loss P _fe becomes larger than when P fe is 0. When the initial phase φ is −π/4 [rad], it can be said that the value of the initial phase φ is such that “the increase due to the harmonic component is greater than the reduction effect of the fundamental wave current I _f1 .” This value may be set as the minimum phase angle of the initial phase φ.

Examining based on FIGS. 14(a) and (b), when the phase angle of the fundamental sine wave g(t) is π/2 radian (rad), the numerical value of the fundamental sine wave g(t) is the signal at that time. The initial phase φ of the fifth harmonic is set to be equal to or higher than the value of the wave h(t), and when it operates, the fundamental wave current I _f1 is lower than when the fundamental sine wave g(t) is used as the signal wave h(t) It is thought that this will reduce the It is also possible that the iron loss P _fe is also reduced.

Explaining using the fundamental sine wave g(t) shown in the above equation (10) and the signal wave h(t) shown in the equation (11), the phase angle (2πf ₁ t) of the fundamental sine wave g(t) is π /2 radian, the value m・sin(π/2) of the fundamental sine wave g(t) is the signal wave h(t)=m・sin(π/2)+n・sin(5・π/2+φ) The initial phase φ is set to be greater than or equal to the value of . In other words, the condition that g(t) (=m・sin(π/2)) ＞= h(t)(=m・sin(π/2)+n・sin(5・π/2+φ)) is satisfied. The initial phase φ of the fifth harmonic is set at .

FIG. 15 is a diagram schematically showing the flow of setting the superimposition ratio n/m in the ring test.

In this ring test, the initial phase φ of the fifth harmonic is set to 0 [rad], and the measurement is performed while changing the superimposition ratio n/m from the larger side to the smaller side.

First, the superimposition ratio n/m is set to the maximum value (step S100). For example, under the measurement conditions shown in FIG. 7(b), the superimposition ratio n/m is set to 0.25, which is the maximum value.

Next, the ring test device 109 is adjusted to the measurement conditions (step S101). For example, under the measurement conditions shown in FIG. 7(b), the ring test device 109 is operated so that the input voltage V _dc to the IGBT inverter 62 is 15 [V], and the fundamental sine wave magnetic flux density B _f1 is 1 [T]. The basic sine wave modulation rate m is adjusted so that Then, a PWM signal is generated by configuring the signal wave h(t) shown in the above equation (11) so that the superimposition ratio is set to n/m, and this PWM signal is input to the IGBT inverter 102 to operate it. . The IGBT inverter 102 outputs a pulse width modulated voltage based on the PWM signal, and this pulse width modulated voltage is applied to the primary coil. This causes a primary current _I1 to flow through the primary coil.

Next, the primary current I ₁ flowing in the primary coil wound around the ring sample 101 of the ring test device 109 and the secondary voltage V ₂ generated in the secondary coil wound around the ring sample 101 are measured (S102 step).

After measuring the primary current _I1 and the secondary voltage _V2 , the superimposition ratio n/m is set to be lowered by a predetermined amount (step S103). For example, in the case of the measurement conditions shown in FIG. 7(b), the superimposition ratio n/m is set to be reduced by 0.05.

It is checked whether the superimposition ratio n/m set in this way is greater than or equal to the minimum value, and if it is greater than or equal to the minimum value (Yes in step S104), the process returns to step S101 and the measurement is repeated. Here, the minimum value of the superimposition ratio n/m used for the inspection is, for example, a value at which the superimposition ratio n/m is −0.25 under the measurement conditions shown in FIG. 7(b).

If the set superimposition ratio n/m is not equal to or greater than the minimum value (No in step S104), the fundamental wave current I _f1 , iron loss P _fe , harmonic current I _{harmonic are} determined for the measurement data of each superposition ratio n/m. is calculated (step S105).

The harmonic current I _harmonic in the ring test is the effective value of the harmonic component (harmonic component of the primary current I ₁ ) of the fundamental sine wave frequency f ₁ of the primary current I ₁ flowing through the primary coil.

Next, the upper limit value of the superimposition ratio n/m is determined (upper limit value determination step, upper limit value determination step) (step S106).

As a method of determining this upper limit value, the upper limit of the range in which the fundamental wave current I _f1 is lower than when the modulation rate n of the fifth harmonic is 0 (zero) is determined by determining the superimposition ratio n/m of the superposition ratio n/m. It may also be an upper limit value. In other words, the upper limit is the superposition ratio n/m, which is the upper limit of the range of the superposition ratio n/m in which the fundamental wave current I _f1 decreases compared to the fundamental wave current I _f1 when the superposition ratio n/m is 0 (zero). The value may be determined. For example, in FIG. 10, the fundamental wave current I _f1 decreases in the range where the superposition ratio n/m is less than 0, compared to when the superposition ratio n/m is 0, so the upper limit of this range is The superimposition ratio n/m becomes less than 0. In this case, the upper limit of the superimposition ratio n/m may be determined to be less than 0. Further, in a range where the superimposition ratio n/m is -0.1 or less, the fundamental wave current I _f1 decreases rapidly, and if this range is used, the upper limit becomes -0.1. In this case, the upper limit of the superimposition ratio n/m is determined to be -0.1. In FIG. 12, the fundamental wave current I _f1 decreases in the range where the superimposition ratio n/m is smaller than 0, compared to when the superimposition ratio n/m is 0, and the upper limit of the superposition ratio n/m is determined to be less than 0. This corresponds to determining the upper limit value of the superimposition ratio n/m to a value of the superimposition ratio n/m at which "the fundamental wave current I _f1 starts to decrease due to a change in magnetic characteristics."

In addition, as a method for determining the upper limit value of the superimposition ratio n/m, compared to the case where the modulation ratio n of the fifth harmonic is 0 (zero), "based on the primary current I ₁ and the secondary voltage V ₂ The upper limit value may be determined as a superimposition ratio n/m that is the upper limit of the range below which the iron loss P _fe as the calculated predetermined loss falls. In other words, the upper limit value is determined to be the superposition ratio n/m, which is the upper limit of the range of the superposition ratio n/m in which the iron loss P _fe is reduced compared to the iron loss P _fe when the superposition ratio n/m is 0. Good too. For example, in FIGS. 11 and 12, the range of the overlap ratio n/m in which the iron loss P _fe is reduced compared to the iron loss P _fe when the overlap ratio n/m is 0 is −0.3 or more and less than 0. Since this is a range, the upper limit of the superimposition ratio n/m in this range is less than 0. In this case, the upper limit of the superimposition ratio n/m is determined to be less than 0. Note that a loss other than the iron loss P _fe may be used as the predetermined loss.

Next, the lower limit value of the superimposition ratio n/m is determined (lower limit value determination step, lower limit value determination step) (S107 step).

As a method for determining this lower limit value, the modulation rate n of the fifth harmonic is 0 (zero) compared to the amount of reduction in the fundamental wave current I _f1 based on the case where the modulation rate n of the fifth harmonic is 0 (zero). The lower limit value may be determined to be the superimposition ratio n/m, which is the lower limit of the range in which the increase amount of the harmonic current I _harmonic is less than the case where the harmonic current I harmonic is (zero). That is, the determination may be made by comparing the amount of reduction in the fundamental wave current I _f1 and the amount of increase in the harmonic current I _harmonic when the superimposition ratio n/m is 0 (zero) as a reference. . First, using the fundamental wave current I _f1 when the superimposition ratio n/m is 0 (zero) as a reference, the amount of reduction in the fundamental wave current I _f1 from the reference when the superimposition ratio n/m changes is calculated. Further, with the harmonic current I _harmonic when the superposition ratio n/m is 0 (zero) as a reference, the amount of increase in the harmonic current I _harmonic from the reference when the superposition ratio n/m changes is calculated. Then, the amount of reduction of the fundamental wave current I _f1 based on the case where the superimposition ratio n/m is 0 (zero), and the reduction amount of the harmonic current I _harmonic based on the case where the superposition ratio n/m is 0 (zero). Find the range of the superimposition ratio n/m in which the reduction amount of the fundamental wave current I _f1 is lower than the increase amount of the harmonic current I _harmonic by comparing the increase amount, and find the superposition ratio n/m that is the lower limit of this range. The lower limit value of the superimposition ratio n/m may be determined. When the amount of reduction in fundamental wave current I _f1 is lower than the amount of increase in harmonic current I _harmonic , the absolute value of the amount of reduction in fundamental wave current I _f1 becomes larger than the absolute value of the amount of increase in harmonic current I _harmonic . .

In addition, as a method for determining the lower limit value of the superposition ratio n/m, the lower limit of the range in which the iron loss P _fe as a "predetermined loss" is lower than when the modulation rate n of the fifth harmonic is 0 (zero) is determined. The lower limit may be determined to be a superimposition ratio n/m. In other words, the lower limit value may be determined as the lower limit of the range of the superposition ratio n/m in which the iron loss P _fe is reduced compared to the iron loss P _fe when the superposition ratio n/m is 0 (zero). . For example, in FIGS. 11 and 12, the range of the overlap ratio n/m in which the iron loss P _fe is reduced compared to the iron loss P _fe when the overlap ratio n/m is 0 is −0.3 or more and less than 0. Since this is a range, the lower limit of the overlap ratio n/m for this range is -0.3. In this case, the lower limit value of the superimposition ratio n/m is determined to be -0.3.

Note that the lower limit value may be determined within the range where measurement data exists. In this case, the lower limit value is determined to be -0.25 in FIG. 11, and the lower limit value is determined to be -0.2 in FIG. 12.

The above two methods for determining the lower limit value of the superimposition ratio n/m are to determine the lower limit value to a value of the superposition ratio n/m at which "the increase due to the harmonic component is greater than the reduction effect of the fundamental wave current I _f1 ". It corresponds to doing.

Next, a set value of the superimposition ratio n/m is set (superposition ratio setting step, superposition ratio setting step) (step S108). The signal wave h(t) shown in equation (11) above is within the range between the upper limit value of the superimposition ratio n/m determined in step S106 and the lower limit value of the superimposition ratio n/m determined in step S107. Set the setting value of the superimposition ratio n/m. The motor drive system 1 is designed or manufactured based on the set value of this superimposition ratio n/m. That is, as the setting value of this superimposition ratio n/m, the three-phase signal waves h _u (t), h _v (t ), h _w (t) are set.

Note that although FIG. 15 shows the flow of measuring by changing the superimposition ratio n/m from a large side to a small side, it is also possible to change the superimposition ratio n/m from a small side to a large side. Further, although the flow of changing the superimposition ratio n/m by setting the initial phase φ of the fifth harmonic to 0 [rad] has been shown, it may be performed by setting the initial phase to a value other than 0 [rad].

If the initial phase φ of the fifth harmonic is fixed in advance as 0 [rad], etc., the superimposition ratio n/m is set according to the flow shown in FIG. (t) may be determined. If the initial phase φ of the fifth harmonic is not fixed and is set by changing it, then the set value of the initial phase φ is determined as shown below.

FIG. 16 is a diagram showing a schematic flow of setting the initial phase φ of the fifth harmonic in the ring test.

In this ring test, measurements are performed while keeping the superimposition ratio n/m constant and changing the initial phase φ of the fifth harmonic from the larger side to the smaller side.

First, the superimposition ratio n/m is set within the range of the upper limit value and lower limit value of the superimposition ratio n/m (step S110). For example, the superimposition ratio n/m is set according to the flow for setting the superimposition ratio n/m shown in FIG. Under the measurement conditions shown in FIG. 13, the superimposition ratio n/m is set to -0.2. The initial phase φ of the fifth harmonic is changed while the superimposition ratio is fixed at the set superimposition ratio n/m.

Next, the initial phase φ of the fifth harmonic is set to the maximum value (step S111). For example, under the measurement conditions shown in FIG. 13, the initial phase φ is set to 3π/4 [rad] (=135°), which is the maximum value.

Next, the ring test device 109 is adjusted to the measurement conditions (step S112). For example, under the measurement conditions shown in FIG. 13, the ring test device 109 is set such that the input voltage V _dc to the IGBT inverter 62 is 15 [V], and the fundamental sine wave magnetic flux density B _f1 is 1 [T]. Adjust the basic sine wave modulation rate m. Then, a PWM signal is generated by configuring the signal wave h(t) shown in equation (11) above so that the set superimposition ratio n/m and initial phase φ are obtained, and this PWM signal is input to the IGBT inverter 102. and make it work. A pulse width modulated voltage is output from the IGBT inverter 102, and the application of this pulse width modulated voltage causes a primary current _I1 to flow through the primary coil.

Next, the primary current I ₁ flowing in the primary coil wound around the ring sample 101 of the ring test device 109 and the secondary voltage V ₂ generated in the secondary coil wound around the ring sample 101 are measured (S113 step).

After measuring the primary current _I1 and the secondary voltage _V2 , the initial phase φ of the fifth harmonic is set to be lowered by a predetermined amount (step S114). For example, in the measurement conditions shown in FIG. 13, the initial phase φ is set to be lowered by π/4 [rad] (=45°).

It is checked whether the initial phase φ set in this way is greater than or equal to the minimum value, and if it is greater than or equal to the minimum value (Yes in step S115), the process returns to step S112 and the measurement is repeated. Here, the minimum value of the initial phase φ of the fifth harmonic used for inspection is, for example, under the measurement conditions shown in FIG. 13, the initial phase φ is -3π/4 [rad] (=-135°). This is the value.

If the set initial phase φ is not greater than or equal to the minimum value (No in step S115), the fundamental wave current I _f1 , iron loss P _fe , and harmonic current I _harmonic are calculated for the measurement data of each initial phase φ ( S116 step).

Next, the maximum phase angle of the initial phase φ of the fifth harmonic is determined (maximum phase angle determination step, maximum phase angle determination step) (step S117).

As a method of determining the maximum phase angle of this initial phase φ, the initial phase φ that is the maximum value in the range in which the fundamental wave current I _f1 is lower than when the modulation rate n of the fifth harmonic is 0 (zero) is determined as the initial phase. It may also be the maximum phase angle of φ. In other words, the initial phase φ that is the maximum value of the range of initial phase φ in which the fundamental wave current I _f1 decreases compared to the fundamental wave current I _f1 when the superimposition ratio n/m is 0 (zero) is set to the maximum phase angle. It may be decided. For example, in the case of FIG. 14(a), the fundamental wave current I _f1 decreases in the range where the initial phase φ is 0 [rad] or less, compared to the case where the superimposition ratio n/m shown as a horizontal broken line is 0. Therefore, the maximum value in this range is when the initial phase φ is 0 [rad]. In this case, the maximum phase angle of the initial phase φ is determined to be 0 [rad]. In addition, the plot showing the fundamental wave current I _f1 is obtained by interpolating between the plots where the initial phase φ is π/4 [rad] and 0 [rad] that sandwich the horizontal broken line above and below, and the fundamental wave current I f1 of the horizontal broken line is Alternatively, the initial phase φ that is _f1 may be calculated, and the calculated initial phase φ may be determined as the maximum phase angle. These correspond to determining the maximum phase angle of the initial phase φ of the fifth harmonic to the value of the initial phase φ at which "the fundamental wave current I _f1 starts to decrease due to a change in magnetic characteristics."

In addition, as a method for determining the maximum phase angle of the initial phase φ, the maximum value in the range in which the iron loss P _fe as a "predetermined loss" is lower than that when the modulation rate n of the fifth harmonic is 0 (zero). The maximum phase angle may be determined as the initial phase φ. In other words, the maximum phase angle is determined to be the initial phase φ that is the maximum value in the range of initial phase φ in which the iron loss P _fe is reduced compared to the iron loss P _fe when the superimposition ratio n/m is 0 (zero). You can. For example, compared to the iron loss P _fe when the superposition ratio n/m is 0, which is shown as a horizontal broken line in FIG. 14(b), the range of the initial phase φ in which the iron loss P _fe is reduced is −π/4[ rad] or more and π/2 [rad] or less, the maximum value of the initial phase φ in this range is π/2 [rad]. In this case, the maximum phase angle of the initial phase φ is determined to be π/2 [rad].

Next, the minimum phase angle of the initial phase φ of the fifth harmonic is determined (minimum phase angle determination step, minimum phase angle determination step) (step S118).

As a method of determining the minimum phase angle of this initial phase φ, the amount of reduction of the fundamental wave current I _f1 based on the case where the modulation rate n of the fifth harmonic is 0 (zero) is The minimum phase angle may be determined to be the initial phase φ that is the minimum value within the range in which the amount of increase in the harmonic current I _harmonic is below the case where the modulation rate n is 0 (zero). First, using the fundamental wave current I _f1 when the superimposition ratio n/m is 0 (zero) as a reference, the amount of reduction in the fundamental wave current I _f1 from the reference when the initial phase φ changes is calculated. Furthermore, using the harmonic current I _harmonic when the superimposition ratio n/m is 0 (zero) as a reference, the amount of increase in the harmonic current I _harmonic from the reference when the initial phase φ changes is calculated. Then, the reduction amount of the fundamental wave current I _f1 with reference to the case where the superimposition ratio n/m is 0 (zero), and the reduction amount of the harmonic current I _harmonic with reference to the case where the superposition ratio n/m is 0 (zero). The range of the initial phase φ in which the amount of reduction in the fundamental wave current I _f1 is lower than the amount of increase in the harmonic current I _harmonic is determined by comparing the amount of increase, and the initial phase φ that is the minimum value in this range is set as the initial phase φ. Alternatively, the minimum phase angle may be determined. When the amount of reduction in fundamental wave current I _f1 is lower than the amount of increase in harmonic current I _harmonic , the absolute value of the amount of reduction in fundamental wave current I _f1 is greater than the absolute value of the amount of increase in harmonic current I _harmonic . It gets bigger.

In addition, as a method of determining the minimum phase angle of the initial phase φ, the minimum value in the range where the iron loss P _fe as a "predetermined loss" is lower than that when the modulation rate n of the fifth harmonic is 0 (zero). The initial phase φ may be determined to be the minimum value. That is, the minimum phase angle may be determined as the minimum value of the range of initial phase φ in which the iron loss P _fe is reduced compared to the iron loss P _fe when the superimposition ratio n/m is 0 (zero). . For example, compared to the iron loss P _fe when the superposition ratio n/m is 0, which is shown as a horizontal broken line in FIG. 14(b), the range of the initial phase φ in which the iron loss P _fe is reduced is −π/4[ rad] or more and π/2 [rad] or less, the minimum value of the initial phase φ in this range is −π/4 [rad]. In this case, the minimum phase angle of the initial phase φ is determined to be −π/4 [rad].

The above two methods for determining the minimum phase angle of the initial phase φ are to determine the minimum phase angle to a value of the initial phase φ at which “the increase due to the harmonic component is greater than the reduction effect of the fundamental wave current I _f1 ”. corresponds to

Next, a set value of the initial phase φ is set (initial phase setting step, initial phase setting step) (step S119). In the range between the maximum phase angle of the initial phase φ determined in step S117 and the minimum phase angle of the initial phase φ determined in step S118, the 5th order of the signal wave h(t) shown in the above equation (11) Set the setting value of the initial phase φ of the harmonic. The motor drive system 1 is designed or manufactured based on the set value of this initial phase φ. That is, as the setting value of this initial phase φ, the three-phase signal waves h _u (t), h _v (t), h _w (t) is set.

Although FIG. 16 shows the flow of measuring by changing the initial phase φ from the larger side to the smaller side, the initial phase φ may also be changed from the smaller side to the larger side.

Also, as the setting value of the initial phase φ of the fifth harmonic of the signal wave h(t) shown in the above equation (11), the phase angle of the fundamental sine wave g(t) shown in the above equation (10) is π/2 The initial phase φ of the fifth harmonic may be set such that the value of the fundamental sine wave g(t) in radians is greater than or equal to the value of the signal wave h(t) at that time. That is, when the phase angle (2πf ₁ t) of the fundamental sine wave g(t) is π/2 radian, the numerical value m・sin(π/2) of the fundamental sine wave g(t) is the signal wave h(t) The initial phase φ may be set to be equal to or greater than the value of =m·sin(π/2)+n·sin(5·π/2+φ).

<Setting method, design method, and manufacturing method by motor test>
Next, a method of setting, designing, and manufacturing the motor drive system 1 through a motor test will be described.

FIG. 17 is a schematic configuration diagram of a motor testing device 289 used for setting, designing, and manufacturing this motor drive system 1. This motor test device 289 has the same configuration as the motor test device 89 used in characteristic evaluation test 2 (motor test) described later (therefore, refer to characteristic evaluation test 2 (motor test) described later for details).

FIG. 18 is a diagram showing an example of a test motor (embedded permanent magnet synchronous motor 273) for the motor test, in which (a) is a schematic cross-sectional view of the motor, and (b) is a diagram showing the specifications of the motor. The same motor as the permanent magnet synchronous motor 5 used in the motor drive system 1 is used as the embedded structure permanent magnet synchronous motor 273 in this motor test. Furthermore, this test motor (embedded structure permanent magnet synchronous motor 273) is composed of a rotor and a stator, and the iron core material of the rotor and stator is the same as the ring sample 101 of the ring test device 109 used in the above ring test method. It is the material of

The IGBT inverter 271 shown in FIG. 17 is a three-phase Si-IGBT inverter equipped with a Si-IGBT as a switching element and a Si diode as a freewheeling diode. This IGBT inverter 271 has the same configuration as the three-phase inverter section 2 of the motor drive system 1, and the switching elements and free wheel diodes used in this IGBT inverter 271 are the same as the three-phase inverter section 2 of the motor drive system 1. The switching elements S ₁ to S ₆ and free wheel diodes D ₁ to D ₆ used are the same.

A fifth harmonic superimposition PWM method is adopted for controlling this IGBT inverter 271. That is, a PWM signal is generated by switching the pulse width at the intersection of the signal wave h(t) on which the fifth harmonic shown in the above equation (11) is superimposed and the carrier wave configured as a triangular wave, and this PWM signal is It is input to the IGBT inverter 271. The IGBT inverter 271 to which the PWM signal is input outputs a PWM drive voltage that rotates the embedded permanent magnet synchronous motor 273, causing the synchronous motor 273 to rotate.

The power measurement and waveform observation are performed using the power meter 272.

FIG. 19 is a diagram showing an example of measurement conditions (superimposition ratio) for a motor test. The measurement conditions are, for example, rotational speed ω of 750 [rpm], average torque T of 0.611 [Nm], carrier frequency _fc of 1 [kHz], and input voltage V _dc to the IGBT inverter 271 of 50 [V]. ], and the basic sine wave modulation rate m is adjusted by performing feedback control so that the rotational speed ω and the torque T are constant. In addition, in the example of the measurement conditions shown in FIG. 19, the range to be measured by changing the superimposition ratio n/m is from −0.25 to 0, but the overlap ratio n/m becomes positive. The measurement may be performed by changing the superimposition ratio over a wider range of n/m on the plus side and the minus side. Note that in FIG. 19, the initial phase φ of the fifth harmonic is 0 [rad].

Next, a loss calculation method of the motor testing device 289 will be explained. Input power P _in to the IGBT inverter 271, input power P _u , P _v , P _w of each phase of the motor, effective input current value I _{u_rms} , I _{v_rms} , I _{w_rms} , input current I _u of each phase of the motor , I _v , and I _w are measured and used to calculate the loss.

The input current effective values I _{u_rms} , I _{v_rms} , and I _{w_rms} of each phase of the motor are measured and calculated as the effective values of the input currents I _u , I _v , and I _w of each phase of the motor.

As shown in equation (17), the overall loss P _total of this motor testing device 289 is composed of an inverter loss P _inv , a copper loss P _Cu , and a motor core loss/mechanical loss P _core&mech . The inverter loss P _inv is the input power P _in to the IGBT inverter 271 , the input power P _u , P _v , P _w of each phase of the motor, the loss P _{w of the power measuring device 272 .} It is calculated using equation (18) using _m . Loss P of power meter 272 _{w. m} is determined by the shunt resistance R _shunt (=0.1 [Ω]), the resistance R _cable (=0.012 [Ω]) of the connection cable, and the effective input current values of each phase of the motor I _{u_rms} , I _{v_rms} , I _{w_rms} Calculate as shown in equation (19). The copper loss P _Cu is calculated as shown in equation (20) using the winding resistance R (=0.5 [Ω]) and the effective input current values I _{u_rms} , I _{v_rms} , and I _{w_rms} of each phase of the motor. The motor core loss/mechanical loss P _core&mech is calculated as shown in equation (21) using the input power P _u , P _v , P _w of each phase of the motor, the copper loss P _Cu , and the mechanical output ωT. Note that since it is difficult to directly measure both motor core loss P _core and mechanical loss P _mech , mechanical loss P _mech and motor core loss P _core are not classified, and the loss is calculated as motor core loss/mechanical loss P _{core & mech.} did.
Examples of measurement results in motor tests measured while changing the superimposition ratio n/m are shown below.

FIG. 20 is a diagram showing an example of the fundamental wave current I _{f1_rms} and the fundamental sine wave modulation rate m obtained by motor testing.

The fundamental wave current I _{f1_rms} in the motor test is the phase average of the effective value of the component of the fundamental sine wave frequency f ₁ calculated from the input currents I _u , I _v , I _w of each phase of the motor.

This figure shows the fifth order of the fundamental sine wave modulation rate m obtained by feedback control so that the rotational speed ω (=750 [rpm]) and the average torque T (=0.611 [Nm]) are constant. It shows the superimposition rate characteristics of harmonics. The fundamental sine wave modulation rate m is -0.25, -0.15, -0. It is decreasing by 1. Also, the fundamental wave current I _{f1_rms} has a superimposition rate n/m of -0.25, -0.15, -0 compared to the case where the fifth harmonic is not superimposed (the case where the superposition ratio n/m is 0). It has decreased by .1. This is the same phenomenon as the measurement result of the ring test described above (see FIG. 10). When the superposition ratio n/m is less than 0, it can be said that the value of the superposition ratio n/m is such that "the fundamental wave current I _{f1_rms} starts to decrease due to a change in the magnetic characteristics." This value may be set as the upper limit of the superimposition ratio n/m.

FIG. 21 is a diagram illustrating an example of the total loss P _total due to a motor test. The horizontal broken line in the figure indicates the measured value of the total loss P _total when the modulation rate n of the fifth harmonic is 0, that is, when the fundamental sine wave g(t) is used as the signal wave h(t). Showing.

The overall loss P _total is greater when the superposition rate n/m is -0.25, -0.15, and -0.1 compared to the case where the fifth harmonic is not superimposed (when the superposition ratio n/m is 0). is decreasing. Further, when the superimposition ratio n/m is -0.15, it shows the minimum value, and the loss reduction rate is 1.5% compared to the case where the fifth harmonic is not superimposed. Also, when the superimposition ratio n/m is −0.10, the loss reduction rate is 1.4%, and the loss is greatly reduced. From this figure, it is assumed that when the superimposition ratio n/m becomes less than -0.25, the overall loss P _total becomes larger than when the fifth harmonic is not superimposed. Because of this assumption, when the superposition ratio n/m is -0.25, the value of the superposition ratio n/m will be such that "the increase due to the harmonic component is greater than the reduction effect of the fundamental wave current I _{f1_rms} ". I can say that. This value may be set as the lower limit of the superimposition ratio n/m.

Below, examples of measurement results in motor tests measured while changing the initial phase φ of the fifth harmonic are shown.

FIG. 22 is a diagram showing an example of measurement conditions (phase angle) for a motor test in which measurement is performed by changing the initial phase φ of the fifth harmonic. In this example of measurement conditions, the range to be measured by changing the initial phase φ is only 0 [rad] and π/4 [rad], but the range where the initial phase φ is negative includes the range where the initial phase φ is negative. Measurement may be performed by changing the initial phase φ within a wider range between the negative side and the negative side. Note that in FIG. 22, the superimposition ratio n/m is −0.1, which is 0.

FIG. 23 is a diagram showing an example of the fundamental sine wave modulation factor m and the fundamental wave current I _{f1_rms} in a motor test. On the horizontal axis, the left side is when the superposition rate n/m is 0, the center is when the superposition rate n/m is -0.1 and the initial phase φ of the fifth harmonic is 0 [rad], and the right side is the superposition rate n/m. The case where m is −0.1 and the initial phase φ is π/4 [rad] is shown.

This figure shows the basic sine wave modulation rate m when feedback control is performed so that the rotation speed ω (=750 [rpm]) and the average torque T (=0.611 [Nm]) are constant. Both the fundamental sine wave modulation rate m and the fundamental wave current I _{f1_rms} are higher when the superposition ratio n/m is -0.1 than when the fifth harmonic is not superimposed (when the superposition ratio n/m is 0). is decreasing. In addition, under the condition that the superimposition ratio n/m is -0.1, the initial phase φ of the fifth harmonic is π/4 [rad] than 0 [rad], and the fundamental sine wave modulation rate m and the fundamental wave current I _{f1_rms} is decreasing.

If the data of the initial phase φ is only at two points, 0 [rad] and π/4 [rad], and if we consider this range, if the initial phase φ becomes smaller than π/4 [rad], "change in magnetic properties" will occur. It can be said that the fundamental wave current I _{f1_rms} starts to decrease. The value of this initial phase φ (π/4) may be set as the maximum phase angle.

FIG. 24 is a diagram illustrating an example of the total loss P _total due to a motor test. The horizontal broken line in the figure indicates the measured value of the overall loss P _total when the fundamental sine wave g(t) is used as the signal wave h(t) (when the modulation rate n of the fifth harmonic is 0). ing.

The overall loss P _total is reduced when the superposition ratio n/m is -0.1 compared to the case where the fifth harmonic is not superimposed (the case where the superposition ratio n/m is 0). Further, under the condition that the superposition ratio n/m is −0.1, the overall loss P _total is reduced when the initial phase φ of the fifth harmonic is 0 [rad] than π/4 [rad].

When the initial phase φ is decreased from π/4 [rad], the overall loss P _total decreases when the initial phase φ is 0. However, it is expected that the overall loss P _total will increase as the initial phase φ is further reduced. Then, the initial phase φ is further reduced, and when the initial phase _φ has a value that is almost the same as the total loss P _total when the superimposition ratio n/m is 0, “the fundamental wave current I _{f1_rms} This can be said to be the value of the initial phase φ where the increase due to harmonic components is greater than the reduction effect. This value may be set as the minimum phase angle of the initial phase φ.

Examining based on FIGS. 23 and 24, the numerical value of the fundamental sine wave g(t) when the phase angle of the fundamental sine wave g(t) is π/2 radian as shown in FIG. 3(a) is The initial phase φ of the 5th harmonic is set to be greater than the value of the signal wave h(t) at that time, and when it operates, compared to the case where the 5th harmonic is not superimposed (when the superimposition rate n/m is 0) Therefore, it is conceivable that the fundamental wave current I _{f1_rms} decreases. It is also possible to reduce the overall loss P _total .

For example, when explaining using the fundamental sine wave g _u (t) shown in the above equation (1) and the signal wave h _u (t) shown in the above equation (2), the phase angle of the fundamental sine wave g _u (t) ( The numerical value m・sin(π/2) of the fundamental sine wave g _u (t) when 2πf ₁ t) is π/2 radian is the signal wave h _u (t)=m・sin(π/2)+n・The initial phase φ is set to be equal to or greater than the value of sin(5·π/2+φ). In other words, the condition that g _u (t) (= m・sin (π/2)) ＞= h _u (t) (= m・sin (π/2) + n・sin (5・π/2 + φ)) is The initial phase φ of the fifth harmonic is set so as to satisfy the following conditions.

FIG. 25 is a diagram schematically showing the flow of setting the superimposition ratio n/m in a motor test.

In this motor test, the initial phase φ of the fifth harmonic is set to 0 [rad], and measurements are performed while changing the superimposition ratio n/m from a large side to a small side.

First, the superimposition ratio n/m is set to the maximum value (step S200). For example, when changing the superimposition ratio n/m in the range of −0.3 or more and 0.3 or less, the superposition ratio n/m is set to 0.3, which is the maximum value.

Next, the motor testing device 289 is adjusted to the measurement conditions (step S201). For example, under the measurement conditions shown in FIG. 19, the motor testing device 289 is configured such that the rotational speed ω is 750 [rpm], the average torque T is 0.611 [Nm], the carrier frequency _fc is 1 [kHz], and the IGBT inverter The basic sine wave modulation rate m is adjusted by setting the input voltage V _dc to 271 to 50 [V] and performing feedback control so that the rotational speed ω and the torque T are constant. Then, a PWM signal is generated by configuring the signal wave h(t) shown in the above equation (11) so that the set superimposition ratio n/m is achieved, and this PWM signal is input to the IGBT inverter 271. The IGBT inverter 271 outputs a PWM drive voltage based on the PWM signal, and the embedded permanent magnet synchronous motor 273 is driven to rotate.

Next, the input power P _in to the IGBT inverter 271 of the motor test device 289, the input power P _u , P _v , P _w of each phase of the motor, the effective input current value of each phase of the motor I _{u_rms} , I _{v_rms} , I _{w_rms} , The input currents I _u , I _v , I _w of each phase of the motor are measured (step S202).

After measuring the input power P _in, etc., the superimposition ratio n/m is set to be lowered by a predetermined amount (step S203). For example, when the predetermined amount is 0.05, the superimposition ratio n/m is set to be reduced by 0.05.

It is checked whether the superimposition ratio n/m set in this way is greater than or equal to the minimum value, and if it is greater than or equal to the minimum value (Yes in step S204), the process returns to step S201 and the measurement is repeated. Here, the minimum value of the superposition ratio n/m used for inspection means, for example, when the superposition ratio n/m is changed in the range of -0.3 or more and 0.3 or less, the superposition ratio n/m is -0.3.

If the set superimposition ratio n/m is not greater than or equal to the minimum value (No in step S204), for the measurement data of each superimposition ratio n/m, the phase average fundamental wave current I _{f1_rms} , motor core loss/mechanical loss P _core&mech Each loss P, total loss P _total , and phase average harmonic current I _{harmonic_rms} are calculated (step S205).

The harmonic current I _{harmonic_rms} in the motor test is the harmonic component of the fundamental sine wave frequency f ₁ calculated from the input current I _u , I _v , I _w of each phase of the motor (the harmonic component of the input current I _u , I _v , I _w is the phase average of the effective values of harmonic components).

Next, the upper limit value of the superimposition ratio n/m is determined (upper limit value determination step, upper limit value determination step) (step S206).

As a method of determining this upper limit value, the upper limit of the range in which the fundamental wave current I _{f1_rms} is lower than when the modulation rate n of the fifth harmonic is 0 (zero) is determined by determining the superimposition ratio n/m of the superposition ratio n/m. It may also be an upper limit value. In other words, the upper limit is the superposition ratio n/m, which is the upper limit of the range of the superposition ratio n/m in which the fundamental wave current I _{f1_rms} decreases compared to the fundamental wave current I _{f1_rms} when the superposition ratio n/m is 0 (zero). The value may be determined. For example, in FIG. 20, the fundamental wave current I _{f1_rms} decreases in a range where the superimposition ratio n/m is less than 0, compared to when the superimposition ratio n/m is 0, so the upper limit of this range is The superimposition ratio n/m becomes less than 0. In this case, the upper limit of the superimposition ratio n/m may be determined to be less than 0. This corresponds to determining the upper limit value of the superimposition ratio n/m to a value of the superimposition ratio n/m at which "the fundamental wave current I _{f1_rms} starts to decrease due to a change in magnetic characteristics."

In addition, as a method for determining the upper limit value of the superimposition ratio n/m, compared to the case where the modulation ratio n of the fifth harmonic is 0 (zero), "input power P _in to the IGBT inverter 271, synchronous motor each phase The total loss P as a predetermined loss calculated based on at least one of the input powers P _u , P _v , P _w of the synchronous motor, or the input currents I _u , I _v , I _w of each phase of the synchronous motor. The upper limit value may be determined as the superimposition ratio n/m that is the upper limit of the range below which _{the total} falls. That is, the upper limit value is determined to be the superposition ratio n/m, which is the upper limit of the range of the superposition ratio n/m in which the overall loss P _total is reduced compared to the total loss P _total when the superposition ratio n/m is 0. Good too. For example, in FIG. 21, the range of the superposition ratio n/m in which the overall loss P _total is reduced compared to the total loss P _total when the superposition ratio n/m is 0 is a range of -0.25 or more and less than 0. Therefore, the upper limit of the superimposition ratio n/m in this range is less than 0. In this case, the upper limit of the superimposition ratio n/m is determined to be less than 0. Note that other losses other than the overall loss P _total , such as motor core loss/mechanical loss P _core&mech , inverter loss P _inv , and copper loss P _Cu , may be used as the predetermined loss.

Next, the lower limit value of the superimposition ratio n/m is determined (lower limit value determination step, lower limit value determination step) (step S207).

As a method for determining this lower limit value, the modulation rate n of the fifth harmonic is 0 (zero) compared to the amount of reduction in the fundamental wave current I _{f1_rms} based on the case where the modulation rate n of the fifth harmonic is 0 (zero). The lower limit value may be determined to be the superimposition ratio n/m, which is the lower limit of the range in which the amount of increase in the harmonic current I _{harmonic_rms} is lower than the case where the harmonic current I harmonic_rms is zero. That is, the determination may be made by comparing the amount of reduction in the fundamental wave current I _{f1_rms} and the amount of increase in the harmonic current I _{harmonic_rms} when the superimposition ratio n/m is 0 (zero). . First, using the fundamental wave current I _{f1_rms} when the superimposition ratio n/m is 0 (zero) as a reference, the amount of reduction in the fundamental wave current I _{f1_rms} from the reference when the superimposition ratio n/m changes is calculated. Furthermore, with the harmonic current I _{harmonic_rms} when the superposition ratio n/m is 0 (zero) as a reference, the amount of increase in the harmonic current I _{harmonic_rms} from the reference when the superposition ratio n/m changes is calculated. Then, the amount of reduction of the fundamental wave current I _{f1_rms} with reference to the case where the superimposition ratio n/m is 0 (zero), and the amount of reduction of the harmonic current I _{harmonic_rms} with reference to the case where the superposition ratio n/m is 0 (zero). Find the range of the superimposition ratio n/m in which the reduction amount of the fundamental wave current I _{f1_rms} is lower than the increase amount of the harmonic current I _{harmonic_rms} by comparing the increase amount, and find the superimposition ratio n/m that is the lower limit of this range. The lower limit value of the superimposition ratio n/m may be determined. In a state where the amount of reduction in the fundamental wave current I _{f1_rms} is lower than the amount of increase in the harmonic current I _{harmonic_rms} , the absolute value of the amount of reduction in the fundamental wave current I _{f1_rms} is larger than the absolute value of the amount of increase in the harmonic current I _{harmonic_rms} .

In addition, as a method for determining the lower limit value of the superposition ratio n/m, the lower limit of the range in which the overall loss P _total as a "predetermined loss" is lower than when the modulation rate n of the fifth harmonic is 0 (zero) is determined. The lower limit may be determined to be a superimposition ratio n/m. That is, the lower limit value may be determined as the lower limit of the range of the superposition ratio n/m in which the overall loss P _total is reduced compared to the total loss P _total when the superposition ratio n/m is 0 (zero). . For example, in FIG. 21, the range of the superposition ratio n/m in which the overall loss P _total is reduced compared to the total loss P _total when the superposition ratio n/m is 0 is a range of -0.25 or more and less than 0. Therefore, the lower limit of the superimposition ratio n/m in this range is -0.25. In this case, the lower limit value of the superimposition ratio n/m is determined to be -0.25.

The above two methods for determining the lower limit value of the superposition ratio n/m are to determine the lower limit value to a value of the superposition ratio n/m at which "the increase due to the harmonic component is greater than the reduction effect of the fundamental wave current I _{f1_rms} " It corresponds to doing.

Next, a set value of the superimposition ratio n/m is set (superposition ratio setting step, superposition ratio setting step) (step S208). The signal wave h(t) shown in equation (11) above is within the range between the upper limit value of the superimposition ratio n/m determined in step S206 and the lower limit value of the superimposition ratio n/m determined in step S207. Set the setting value of the superimposition ratio n/m. The motor drive system 1 is designed or manufactured based on the set value of this superimposition ratio n/m. That is, as the setting value of this superimposition ratio n/m, the three-phase signal waves h _u (t), h _v (t ), h _w (t) are set.

Although FIG. 25 shows the flow of measuring by changing the superimposition ratio n/m from a large side to a small side, it is also possible to change the superimposition ratio n/m from a small side to a large side. Further, although the flow of changing the superimposition ratio n/m by setting the initial phase φ of the fifth harmonic to 0 [rad] has been shown, it may be performed by setting the initial phase to a value other than 0 [rad].

When the initial phase φ of the fifth harmonic is fixed in advance as 0 [rad], etc., the superimposition ratio n/m is set according to the flow shown in FIG. 25 to generate the signal wave h shown in the above equation (11). (t) may be determined. If the initial phase φ of the fifth harmonic is not fixed and is set by changing it, then the set value of the initial phase φ is determined as shown below.

FIG. 26 is a diagram showing a schematic flow of setting the initial phase φ of the fifth harmonic in a motor test.

In this motor test, the superimposition ratio n/m is kept constant and the initial phase φ of the fifth harmonic is changed from the larger side to the smaller side.

First, the superimposition ratio n/m is set within the range of the upper limit value and the lower limit value of the superimposition ratio n/m (step S210). For example, the superimposition ratio n/m is set according to the flow for setting the superimposition ratio n/m shown in FIG. Under the measurement conditions shown in FIG. 22, the superimposition ratio n/m is set to -0.1. The initial phase φ of the fifth harmonic is changed while the superimposition ratio is fixed at the set superimposition ratio n/m.

Next, the initial phase φ of the fifth harmonic is set to the maximum value (step S211). For example, when changing the initial phase φ in the range from -3π/4[rad] (=-135°) to 3π/4[rad] (=135°), the maximum value is 3π/4[rad] (=-135°). rad] (=135°).

Next, the motor testing device 289 is adjusted to the measurement conditions (step S212). For example, under the measurement conditions shown in FIG. 22, the motor testing device 289 is configured such that the rotational speed ω is 750 [rpm], the average torque T is 0.611 [Nm], the carrier frequency _fc is 1 [kHz], and the IGBT inverter The basic sine wave modulation rate m is adjusted by setting the input voltage V _dc to 271 to 50 [V] and performing feedback control so that the rotational speed ω and the torque T are constant. Then, a PWM signal is generated by configuring the signal wave h(t) shown in the above equation (11) so that the set superimposition ratio n/m is achieved, and this PWM signal is input to the IGBT inverter 271. The IGBT inverter 271 outputs a PWM drive voltage based on the PWM signal, and the embedded permanent magnet synchronous motor 273 is driven to rotate.

Next, the input power P _in to the IGBT inverter 271 of the motor test device 289, the input power P _u , P _v , P _w of each phase of the motor, the effective input current value of each phase of the motor I _{u_rms} , I _{v_rms} , I _{w_rms} , The input currents I _u , I _v , I _w of each phase of the motor are measured (step S213).

After measuring the input power P _in , etc., the initial phase φ of the fifth harmonic is set by lowering it by a predetermined amount (step S214). For example, if the predetermined amount is π/4 [rad] (=45°), the initial phase φ is set to be lowered by π/4 [rad] (=45°).

It is checked whether the initial phase φ set in this way is greater than or equal to the minimum value, and if it is greater than or equal to the minimum value (Yes in step S215), the process returns to step S212 and the measurement is repeated. Here, the minimum value of the initial phase φ of the fifth harmonic used for inspection means, for example, that the initial phase φ is -3π/4 [rad] (=-135°) or more and 3π/4 [rad] (= 135°), the initial phase φ is −3π/4 (=−135°).

If the set initial phase φ is not greater than or equal to the minimum value (No in step S215), each loss such as the phase average fundamental wave current I _{f1_rms} and motor core loss/mechanical loss P _core&mech is calculated for the measurement data of each initial phase φ. P, the overall loss P _total , and the phase average harmonic current I _{harmonic_rms} are calculated (step S216).

Next, the maximum phase angle of the initial phase φ of the fifth harmonic is determined (maximum phase angle determination step, maximum phase angle determination step) (S217 step).

As a method of determining the maximum phase angle of this initial phase φ, the initial phase φ that is the maximum value in the range where the fundamental wave current I _{f1_rms} is lower than when the modulation rate n of the fifth harmonic is 0 (zero) is determined as the initial phase. It may also be the maximum phase angle of φ. In other words, the initial phase φ, which is the maximum value in the range of initial phase φ in which the fundamental wave current I _f1 decreases compared to the fundamental wave current I _{f1_rms} when the superposition ratio n/m is 0 (zero), is set to the maximum phase angle. It may be decided. This corresponds to determining the maximum phase angle of the initial phase φ to a value of the initial phase φ at which “the fundamental wave current I _{f1_rms} starts to decrease due to a change in magnetic characteristics.”

In addition, as a method for determining the maximum phase angle of the initial phase φ, the maximum value in the range where the overall loss P _total as a "predetermined loss" is lower than when the modulation rate n of the fifth harmonic is 0 (zero). The maximum phase angle may be determined as the initial phase φ. In other words, the maximum phase angle is determined to be the initial phase φ that is the maximum value in the range of initial phases φ in which the total loss P _total is reduced compared to the total loss P _total when the superposition ratio n/m is 0 (zero). You can.

Next, the minimum phase angle of the initial phase φ of the fifth harmonic is determined (minimum phase angle determination step, minimum phase angle determination step) (S218 step).

As a method of determining the minimum phase angle of this initial phase φ, the amount of reduction of the fundamental wave current I _{f1_rms} based on the case where the modulation rate n of the fifth harmonic is 0 (zero) is The minimum phase angle may be determined to be the initial phase φ that is the minimum value within the range in which the amount of increase in the harmonic current I _{harmonic_rms} is below the case where the modulation rate n is 0 (zero). First, using the fundamental wave current I _{f1_rms} when the superimposition ratio n/m is 0 (zero) as a reference, the amount of reduction in the fundamental wave current I _{f1_rms} from the reference when the initial phase φ changes is calculated. Furthermore, using the harmonic current I _{harmonic_rms} when the superimposition ratio n/m is 0 (zero) as a reference, the amount of increase in the harmonic current I _{harmonic_rms} from the reference when the initial phase φ changes is calculated. Then, the reduction amount of the fundamental wave current I _{f1_rms} based on the case where the superimposition ratio n/m is 0 (zero), and the reduction amount of the harmonic current I _harmonic based on the case where the superposition ratio n/m is 0 (zero). By comparing the amount of increase, find the range of initial phase φ in which the amount of reduction of fundamental wave current I _{f1_rms} is lower than the amount of increase of harmonic current I _{harmonic_rms} , and set the initial phase φ that is the minimum value of this range as initial phase φ The minimum phase angle may be determined. When the amount of reduction in fundamental wave current I _{f1_rms} is lower than the amount of increase in harmonic current I _{harmonic_rms} , the absolute value of the amount of reduction in fundamental wave current I _{f1_rms} is greater than the absolute value of the amount of increase in harmonic current I _{harmonic_rms} . It gets bigger.

In addition, as a method of determining the minimum phase angle of the initial phase φ, the minimum value in the range where the overall loss P _total as a "predetermined loss" is lower than that when the modulation rate n of the fifth harmonic is 0 (zero). The initial phase φ may be determined as the minimum phase angle. That is, the minimum phase angle may be determined as the minimum value of the range of the initial phase φ in which the total loss P _total is reduced compared to the total loss P _total when the superimposition ratio n/m is 0 (zero). .

The above two methods for determining the minimum phase angle of the initial phase φ are to determine the minimum phase angle to a value of the initial phase φ at which “the increase due to the harmonic component is greater than the reduction effect of the fundamental wave current I _{f1_rms} ”. corresponds to

Next, a set value of the initial phase φ is set (initial phase setting step, initial phase setting process) (step S219). In the range between the maximum phase angle of the initial phase φ determined in step S217 and the minimum phase angle of the initial phase φ determined in step S218, the 5th order of the signal wave h(t) shown in equation (11) above Set the setting value of the initial phase φ of the harmonic. The motor drive system 1 is designed or manufactured based on the set value of this initial phase φ. That is, as the setting value of this initial phase φ, the three-phase signal waves h _u (t), h _v (t), h _w (t) is set.

Although FIG. 26 shows the flow of measuring by changing the initial phase φ from the larger side to the smaller side, the initial phase φ may also be changed from the smaller side to the larger side.

Also, as the setting value of the initial phase φ of the fifth harmonic of the signal wave h(t) shown in the above equation (11), the phase angle of the fundamental sine wave g(t) shown in the above equation (10) is π/2 The initial phase φ of the fifth harmonic may be set such that the value of the fundamental sine wave g(t) in radians is greater than or equal to the value of the signal wave h(t) at that time.

The above is a method of setting, designing, and manufacturing this motor drive system 1, in which the signal shown in equation (11) above is obtained by superimposing the fifth harmonic of the fundamental sine wave g(t) shown in equation (10) above. The upper and lower limits of the superimposition rate n/m of wave h(t), the setting value of the superimposition ratio n/m, the maximum and minimum phase angles of the initial phase φ of the fifth harmonic, and the setting value of the initial phase φ We explained how to determine the

The harmonics to be superimposed on the fundamental sine wave g(t) shown in the above formula (10) are not limited to the fifth harmonic, but may be superimposed with the a-th harmonic, where a is an integer greater than or equal to the fifth harmonic. . In this case, the signal wave h _a (t) is expressed as equation (22).
The upper and lower limits of the superimposition ratio n _a /m of the signal wave h _a (t) shown in equation (22) above with the a-th harmonic superimposed using the ring test method or the motor test method described above. , the setting value of the superimposition ratio n _a /m, the maximum phase angle and minimum phase angle of the initial phase φ _a of the a-th harmonic, and the setting value of the initial phase φ _a .

The upper and lower limits of the superimposition ratio n _a /m of the signal wave h _a (t) determined in this way, the set value of the superimposition ratio n _a /m, and the maximum phase angle of the initial phase φ _a of the a-th harmonic. The motor drive system 1 is designed or manufactured based on the set values of the minimum phase angle and the initial phase _φa . That is, the three-phase signal waves h _ua (t), h _va (t), h _wa (t) when the a-th harmonics of the motor drive system 1 shown in the above formulas (7) to (9) are superimposed. is set.

Next, the effects of the motor drive system 1 according to the embodiment of the present invention will be explained.

According to this embodiment, the fifth harmonic is superimposed on the signal wave h(t), and the upper limit of the superimposition ratio n/m is n/m where the fundamental wave currents I _f1 and I _{f1_rms} start to decrease due to changes in magnetic characteristics. It operates so that the lower limit of n/m is a value of n/m at which the increase due to harmonic components is greater than the reduction effect of the fundamental wave currents I _f1 and I _{f1_rms} . Because of this, the loss of the motor drive system 1 can be reduced compared to the case where the fundamental sine wave g(t) is used as the signal wave h(t).

Further, according to the present embodiment, since the operation is performed in a range where the superimposition ratio n/m is greater than −0.3 and less than 0, it is possible to stably reduce the loss of the motor drive system 1.
Further, according to the present embodiment, when the phase angle (2πf ₁ t) of the fundamental sine wave g(t) is π/2 radian, the numerical value m·sin(π/2) of the fundamental sine wave g(t) is , the initial phase φ of the fifth harmonic is greater than or equal to the value of the signal wave h(t)=m·sin(π/2)+n·sin(5·π/2+φ). Because of this, the loss of the motor drive system 1 can be reduced compared to the case where the fundamental sine wave g(t) is used as the signal wave h(t).
Further, according to the present embodiment, the maximum phase angle of the initial phase φ of the fifth harmonic is the value of the initial phase φ at which the fundamental wave current I _f1 , I _{f1_rms} starts to decrease due to a change in magnetic characteristics, and the value of the initial phase φ of the initial phase φ The minimum phase angle operates at a value of the initial phase φ at which the increase due to the harmonic component is greater than the reduction effect of the fundamental wave current I _f1 , I _{f1_rms} . Because of this, the loss of the motor drive system 1 can be reduced compared to the case where the fundamental sine wave g(t) is used as the signal wave h(t).

Further, according to the present embodiment, the fifth or higher harmonics are superimposed on the signal wave h _a (t), and the upper limit of the superimposition rate n _a /m is the fundamental wave current I _f1 , I This is the value of n _a /m at which _{f1_rms} starts to decrease, and the lower limit of n _a /m is the value of n _a /m at which the increase due to harmonic components is greater than the reduction effect of fundamental wave current I _f1 and I _{f1_rms} . It works like this. Because of this, the loss of the motor drive system 1 can be reduced compared to the case where the fundamental sine wave g(t) is used as the signal wave h _a (t).

Further, according to the present embodiment, when the phase angle (2πf ₁ t) of the fundamental sine wave g(t) is π/2 radian, the numerical value m·sin(π/2) of the fundamental sine wave g(t) is , operates with the initial phase _φ a of the fifth or higher a-order harmonic, which is equal to or greater than the value of the signal wave h _a (t ₎ = m・sin (π/2) + na・sin (a・π/2 + φ _{a )} . . Because of this, the loss of the motor drive system 1 can be reduced compared to the case where the fundamental sine wave g(t) is used as the signal wave _ha (t).

Further, according to the present embodiment, the maximum phase angle of the initial phase φ _a of the fifth or higher a-th harmonic is the same as the initial phase φ _a at which the fundamental wave currents I _f1 and I _{f1_rms} begin to decrease due to changes in magnetic properties. The value of the initial phase φ _a is such that the minimum phase angle of the initial phase φ _a is such that the increase due to the harmonic component is greater than the reduction effect of the fundamental wave current I _f1 , I _{f1_rms} . Because of this, the loss of the motor drive system 1 can be reduced compared to the case where the fundamental sine wave g(t) is used as the signal wave h _a (t).

Further, according to the present embodiment, the fifth harmonic is superimposed on the signal wave h(t), and the value of the superimposition rate n/m at which the fundamental wave currents I _f1 and I _{f1_rms} start to decrease due to changes in magnetic properties is n /m as the upper limit, and a lower limit value determining step where the lower limit of n/m is the value of n/m where the increase due to harmonic components is greater than the reduction _{effect of the fundamental wave currents I f1} _and I f1_rms. The motor drive system 1 is manufactured so that the superimposition ratio n/m is set in a range of not less than the lower limit of n/m and not more than the upper limit of n/m. Because of this, it is possible to manufacture a motor drive system 1 that reduces loss compared to the case where the fundamental sine wave g(t) is used as the signal wave h(t).

Further, according to the present embodiment, the fifth or higher harmonics are superimposed on the signal wave h _a (t), and the fundamental wave currents I _f1 and I _{f1_rms} start to decrease due to changes in the magnetic characteristics, resulting in a superimposition rate n _a /m as the upper limit of n _a /m, and the value of n a /m where the increase due to harmonic components is greater than the reduction effect of the fundamental wave currents I _f1 and I _{f1_rms} is determined as n _a / _m . The motor drive system 1 is manufactured so as to include a lower limit value determining step for setting the lower limit of m, and to set the superimposition ratio n _a _/ m in a range that is greater than or equal to the lower limit of n _a /m and less than or equal to the upper limit of n a /m. . Because of this, it is possible to manufacture a motor drive system 1 that reduces loss compared to the case where the fundamental sine wave g(t) is used as the signal wave h _a (t).

Furthermore, according to the present embodiment, the motor drive system 1 is manufactured with the superimposition ratio n/m set to a range greater than -0.3 and less than 0. Because of this, it is possible to manufacture the motor drive system 1 with stable loss reduction.

Further, according to the present embodiment, the fifth harmonic is superimposed on the signal wave h(t), and the value of the superimposition rate n/m at which the fundamental wave currents I _f1 and I _{f1_rms} start to decrease due to changes in magnetic properties is n /m, and a lower limit determining step, which sets the lower limit of n/m to the value of n/m where the increase due to harmonic components is greater than the reduction effect of the fundamental wave currents I _f1 and I _{f1_rms} . The motor drive system 1 is designed so that the superimposition ratio n/m is set in a range that is greater than or equal to the lower limit of n/m and less than or equal to the upper limit of n/m. Because of this, it is possible to design a motor drive system 1 that reduces loss compared to the case where the fundamental sine wave g(t) is used as the signal wave h(t).

Further, according to the present embodiment, the fifth or higher harmonics are superimposed on the signal wave h _a (t), and the fundamental wave currents I _f1 and I _{f1_rms} start to decrease due to changes in the magnetic characteristics, resulting in a superimposition rate n _a /m as the upper limit of n _a /m; and determining the value of n _a /m at which the increase due to harmonic components is greater than the reduction effect of the fundamental wave currents I _f1 and I _{f1_rms} _. The motor drive system 1 is designed to have a step of determining a lower limit value as the lower limit of m, and to set the superimposition ratio n _a _/ m in a range that is greater than or equal to the lower limit of n a /m and less than or equal to the upper limit of n _a /m. . Because of this, it is possible to design a motor drive system 1 that reduces loss compared to the case where the fundamental sine wave g(t) is used as the signal wave h _a (t).

Furthermore, according to the present embodiment, the motor drive system 1 is designed with the superimposition ratio n/m set in a range greater than -0.3 and less than 0. Because of this, it is possible to design a motor drive system 1 that stably reduces loss.

Further, according to the present embodiment, the upper limit value of the superimposition rate n/m, which is the ratio of the modulation rate n of the fifth harmonic to the modulation rate m of the fundamental sine wave g(t) constituting the signal wave h(t). The lower limit value is determined by a ring test, and the upper limit value of the superposition ratio n/m is the upper limit of the range in which the fundamental wave current I _f1 is lower than when the modulation rate n of the fifth harmonic is zero. The superimposition ratio n/m is determined as the upper limit of the range in which the iron loss P _fe is lower than when the modulation rate n of the fifth harmonic is zero, or the superposition ratio n/m Compared to the amount of reduction in fundamental wave current I _f1 based on the case where the modulation rate n of the fifth harmonic is zero, the lower limit of The superimposition ratio n/m that is the lower limit of the range in which the amount of increase in the wave current I _harmonic is below, or the superimposition ratio n that is the lower limit of the range in which the iron loss P _fe is lower than when the modulation rate n of the fifth harmonic is zero. /m. Because of this, loss can be reduced compared to the case where the fundamental sine wave g(t) is used as the signal wave h(t).

Further, according to the present embodiment, the maximum phase angle and minimum phase angle of the initial phase φ of the fifth harmonic are determined by a ring test performed by changing the initial phase φ, and the maximum phase angle is the initial phase φ that is the maximum value in the range in which the fundamental wave current I _f1 is lower than when the modulation rate n of the fifth harmonic is zero, or The initial phase φ is determined as the maximum value in the range in which the loss P _fe is lower, and the minimum phase angle is compared to the amount of reduction in the fundamental wave current I _f1 based on the case where the modulation rate n of the fifth harmonic is zero. , the initial phase φ that is the minimum value in the range below which the amount of increase in the harmonic current I _harmonic is based on the case where the modulation rate n of the fifth harmonic is zero, or the initial phase φ when the modulation rate n of the fifth harmonic is zero. The minimum phase angle is determined as the initial phase φ that is the minimum value in the range in which the iron loss P _fe is lower than in the case. Because of this, the loss of the motor drive system 1 can be reduced compared to the case where the fundamental sine wave g(t) is used as the signal wave h(t).

Further, according to the present embodiment, the upper limit value of the superimposition rate n/m, which is the ratio of the modulation rate n of the fifth harmonic to the modulation rate m of the fundamental sine wave g(t) constituting the signal wave h(t). The lower limit value is determined by motor tests, and the upper limit value of the superposition ratio n/m is the upper limit of the range in which the fundamental wave current I _{f1_rms} is lower than when the modulation rate n of the fifth harmonic is zero. The superimposition ratio n/m is determined as the upper limit of the range in which the overall loss P _total is lower than when the modulation rate n of the fifth harmonic is zero, and The lower limit value is the reduction amount of the fundamental wave current I f1_rms based on the case where the modulation rate n of the fifth harmonic is zero, compared to the amount of reduction of the fundamental wave current I _{f1_rms} based on the case where the modulation rate n of the fifth harmonic is zero. The superimposition ratio n/m is the lower limit of the range in which the increase in the current I _{harmonic_rms} falls below, or the superimposition ratio n/m is the lower limit of the range in which the overall loss P _total is lower than when the modulation rate n of the fifth harmonic is zero. m. Because of this, loss can be reduced compared to the case where the fundamental sine wave g(t) is used as the signal wave h(t).

Further, according to the present embodiment, the maximum phase angle and minimum phase angle of the initial phase φ of the fifth harmonic are determined by a motor test performed by changing the initial phase φ, and the maximum phase angle is the initial phase φ that is the maximum value in the range in which the fundamental wave current I _{f1_rms} is lower than when the modulation rate n of the fifth harmonic is zero, or the overall value is lower than when the modulation rate n of the fifth harmonic is zero. The initial phase φ is determined as the maximum value within the range in which the loss P _total is below, and the minimum phase angle is compared to the amount of reduction in the fundamental wave current I _{f1_rms} based on the case where the modulation rate n of the fifth harmonic is zero. , the initial phase φ that is the minimum value in the range below which the amount of increase in the harmonic current I _{harmonic_rms} is based on the case where the modulation rate n of the fifth harmonic is zero, or the initial phase φ when the modulation rate n of the fifth harmonic is zero. The initial phase φ is determined to be the minimum value in the range in which the overall loss P _total is lower than in the case of the initial phase φ. Because of this, the loss of the motor drive system 1 can be reduced compared to the case where the fundamental sine wave g(t) is used as the signal wave h(t).

Further, according to the present embodiment, the superimposition rate is the ratio of the modulation rate n _a of the fifth or higher harmonics to the modulation rate m of the fundamental sine wave g(t) constituting the signal wave h _a (t). The upper and lower limits of n _a /m are determined by a ring test, and the upper limit of the superimposition rate n _a /m is more fundamental than when the modulation rate n _a of the a-th harmonic is zero. The superimposition ratio n _a /m is the upper limit of the range in which the wave current I _f1 is lower, or the superimposition ratio n a is the upper limit in the range in which the iron loss P _fe is lower than when the modulation rate n of the a-th harmonic is zero _. /m, and the lower limit value of the superimposition rate _n _a _/ m is determined as follows: When the modulation rate na / _m is the lower limit of the range in which the increase in the harmonic current I _harmonic is lower than the case where the modulation rate n _a of is zero, or when the modulation rate n _a of the a-th harmonic is zero The overlap ratio n _a /m is determined as the lower limit of the range in which the iron loss P _fe is lower than . Because of this, loss can be reduced compared to the case where the fundamental sine wave g(t) is used as the signal wave h _a (t).

Further, according to the present embodiment, the superimposition rate is the ratio of the modulation rate n _a of the fifth or higher harmonics to the modulation rate m of the fundamental sine wave g(t) constituting the signal wave h _a (t). The upper and lower limits of n _a /m are determined by motor tests, and the upper limit of the superimposition rate n _a /m is lower than the fundamental wave when the modulation rate n of the a-th harmonic is zero. The superimposition ratio n _a /m is the upper limit of the range in which the current I _{f1_rms} is lower, or the superimposition ratio n a /m is the upper limit in the range in which the overall loss P _total is lower than when the modulation rate _n _a of the a-th harmonic is zero. m, and the lower limit _of the superimposition rate n _a _/ m is determined as The superimposition rate n _a /m is the lower limit of the range in which the amount of increase in the harmonic current I _{harmonic_rms} is lower than the case where the modulation rate n _a is zero, or the case where the modulation rate n _a of the a-th harmonic is zero. is also determined as the overlap ratio n _a /m, which is the lower limit of the range below which the overall loss P _total is. Because of this, the loss can be reduced compared to the case where the fundamental sine wave g(t) is used as the signal wave h _a (t).

Further, according to the present embodiment, the upper limit value of the superimposition rate n/m, which is the ratio of the modulation rate n of the fifth harmonic to the modulation rate m of the fundamental sine wave g(t) constituting the signal wave h(t). The lower limit value is determined by a ring test, and in the upper limit value determination process, the superposition that is the upper limit of the range in which the fundamental wave current I _f1 is lower than when the modulation rate n of the fifth harmonic is zero is determined. The upper limit value is determined as the ratio n/m, or the superimposition ratio n/m, which is the upper limit of the range in which the iron loss P _fe is lower than when the modulation rate n of the fifth harmonic is zero, and in the lower limit value determination step, , compared to the amount of reduction in the fundamental wave current I _f1 based on the case where the modulation factor n of the fifth harmonic is zero, the harmonic current I _harmonic based on the case where the modulation factor n of the fifth harmonic is zero The lower limit is the superimposition ratio n/m, which is the lower limit of the range in which the increase in the amount of _increase in The value is determined, and in the superimposition rate setting step, the motor drive system 1 is manufactured so that the set value of the superimposition ratio n/m of the signal wave h(t) is set within a range of not less than the lower limit value and not more than the upper limit value. be done. Because of this, it is possible to manufacture the motor drive system 1 with reduced loss compared to the case where the fundamental sine wave g(t) is used as the signal wave h(t).

Further, according to the present embodiment, the upper limit value of the superimposition rate n/m, which is the ratio of the modulation rate n of the fifth harmonic to the modulation rate m of the fundamental sine wave g(t) constituting the signal wave h(t). The lower limit value is determined by a motor test, and in the upper limit value determination process, the superposition that is the upper limit of the range in which the fundamental wave current I _{f1_rms} is lower than when the modulation rate n of the fifth harmonic is zero is determined. The upper limit value is determined as the ratio n/m, or the superposition ratio n/m, which is the upper limit of the range in which the overall loss P _total is lower than when the modulation rate n of the fifth harmonic is zero, and in the lower limit value determination step, , compared to the amount of reduction in the fundamental wave current I _{f1_rms} based on the case where the modulation rate n of the fifth harmonic is zero, the harmonic current I _{harmonic_rms} based on the case where the modulation rate n of the fifth harmonic is zero. The lower limit is the superimposition ratio n/m, which is the lower limit of the _range in which the increase amount of The value is determined, and in the superimposition rate setting step, the motor drive system 1 is manufactured so that the set value of the superimposition ratio n/m of the signal wave h(t) is set within a range of not less than the lower limit value and not more than the upper limit value. be done. Because of this, it is possible to manufacture the motor drive system 1 with reduced loss compared to the case where the fundamental sine wave g(t) is used as the signal wave h(t).

Further, according to the present embodiment, the upper limit value of the superimposition rate n/m, which is the ratio of the modulation rate n of the fifth harmonic to the modulation rate m of the fundamental sine wave g(t) constituting the signal wave h(t). The lower limit value is determined by a ring test, and in the upper limit value determination step, the superposition that is the upper limit of the range in which the fundamental wave current I _f1 is lower than when the modulation rate n of the fifth harmonic is zero is determined. The upper limit value is determined as the ratio n/m, or the superimposition ratio n/m, which is the upper limit of the range in which the iron loss P _fe is lower than when the modulation rate n of the fifth harmonic is zero, and in the lower limit value determination step, , compared to the amount of reduction in the fundamental wave current I _f1 based on the case where the modulation factor n of the fifth harmonic is zero, the harmonic current I _harmonic based on the case where the modulation factor n of the fifth harmonic is zero The lower limit is the superimposition ratio n/m, which is the lower limit of the range in which the increase in the amount of _increase in The value is determined, and in the superimposition ratio setting step, the motor drive system 1 is designed so that the set value of the superimposition ratio n/m of the signal wave h(t) is set within a range of not less than the lower limit value and not more than the upper limit value. be done. Because of this, it is possible to design a motor drive system 1 with reduced loss compared to the case where the fundamental sine wave g(t) is used as the signal wave h(t).

Further, according to the present embodiment, the maximum phase angle and minimum phase angle of the initial phase φ of the fifth harmonic are determined by a ring test performed by changing the initial phase φ, and the maximum phase angle In the determination step, the initial phase φ is determined to be the maximum value in the range in which the fundamental wave current I _f1 is lower than when the modulation rate n of the fifth harmonic is zero, or when the modulation rate n of the fifth harmonic is zero. The maximum phase angle is determined as the initial phase φ that is the maximum value in the range in which the iron loss P _fe is below, and in the minimum phase angle determination step, the fundamental wave current is The initial phase φ that is the minimum value in the range in which the increase in harmonic current I _harmonic is less than the amount of reduction in I _f1 when the modulation rate n of the fifth harmonic is zero, or the fifth harmonic The minimum phase angle is determined as the initial phase φ that is the minimum value in the range in which the iron loss P _fe is lower than when the wave modulation rate n is zero, and in the initial phase setting step, the The motor drive system 1 is designed such that the set value of the initial phase φ of the signal wave h(t) is set within the range. Because of this, it is possible to set and design a motor drive system with reduced loss compared to the case where the fundamental sine wave g(t) is used as the signal wave h(t).

Further, according to the present embodiment, the upper limit value of the superimposition rate n/m, which is the ratio of the modulation rate n of the fifth harmonic to the modulation rate m of the fundamental sine wave g(t) constituting the signal wave h(t). The lower limit value is determined by a motor test, and in the upper limit value determination step, a superimposition value that is the upper limit of the range in which the fundamental wave current I _{f1_rms} is lower than when the modulation rate n of the fifth harmonic is zero is determined. The upper limit value is determined as the ratio n/m, or the superimposition ratio n/m, which is the upper limit of the range in which the overall loss P _total is lower than when the modulation rate n of the fifth harmonic is zero, and in the lower limit value determination step, , compared to the amount of reduction in the fundamental wave current I _{f1_rms} based on the case where the modulation rate n of the fifth harmonic is zero, the harmonic current I _{harmonic_rms} based on the case where the modulation rate n of the fifth harmonic is zero. The lower limit is the superimposition ratio n/m, which is the lower limit of the _range in which the increase amount of The value is determined, and in the superimposition ratio setting step, the motor drive system 1 is designed so that the set value of the superimposition ratio n/m of the signal wave h(t) is set within a range of not less than the lower limit value and not more than the upper limit value. be done. Because of this, it is possible to design a motor drive system 1 with reduced loss compared to the case where the fundamental sine wave g(t) is used as the signal wave h(t).

Further, according to the present embodiment, the maximum phase angle and minimum phase angle of the initial phase φ of the fifth harmonic are determined by a motor test performed by changing the initial phase φ, and the maximum phase angle In the determination step, the initial phase φ is determined to be the maximum value in the range in which the fundamental wave current I _{f1_rms} is lower than when the modulation rate n of the fifth harmonic is zero, or when the modulation rate n of the fifth harmonic is zero. The maximum phase angle is determined as the initial phase φ that is the maximum value in the range below the total loss P _total , and in the minimum phase angle determination step, the fundamental wave current is The initial phase φ, which is the minimum value in the range in which the increase in harmonic current I _{harmonic_rms} based on the case where the modulation rate n of the fifth harmonic is zero, is lower than the reduction in I _{f1_rms} , or the fifth harmonic The minimum phase angle is determined as the initial phase φ that is the minimum value in the range in which the overall loss P _total is lower than when the wave modulation rate n is zero, and in the initial phase setting step, the The motor drive system 1 is designed such that the set value of the initial phase φ of the signal wave h(t) is set within the range. Because of this, it is possible to set and design a motor drive system 1 with reduced loss compared to the case where the fundamental sine wave g(t) is used as the signal wave h(t).

Further, according to the present embodiment, the waveforms obtained by superimposing the fifth harmonic on the fundamental sine waves g _u (t), g _v (t), and g _w (t) are used as the signal waves h _u (t), h _v (t), h _w (t), and three-phase pulse width modulation is performed by switching the pulse width at the intersection of the signal waves h _u (t), h _v (t), h _w (t) and the carrier wave. Generate a three-phase PWM drive signal that forms the drive voltage, and calculate the ratio of the modulation rate n of the fifth harmonic to the modulation rate m of the fundamental sine waves g _u (t), g _v (t), g w (t) _. The superimposition ratio n/m is -0.25 or more and -0.05 or less, and the initial phase φ of the fifth harmonic is -π/4 [rad] or more and π/2 [rad] or less. There is. Because of this, loss can be reduced compared to the case where the fundamental sine waves g _u (t), g _v (t), and g _w (t) are used as signal waves.

Furthermore, according to the present embodiment, since the superimposition ratio n/m is greater than or equal to -0.15 and less than or equal to -0.1, the loss is further reduced.

Furthermore, according to the present embodiment, the initial phase φ of the fifth harmonic is greater than or equal to π/8 [rad] and less than or equal to 5π/16 [rad], thereby improving the effect of reducing loss. Also, the loss is reduced compared to the case where the initial phase φ is 0.

Further, according to the present embodiment, the superimposition ratio n/m is −0.15 or more and −0.1 or less, and the initial phase φ of the fifth harmonic is π/4 [rad]. A stable loss reduction effect can be obtained. Also, the loss is reduced compared to the case where the initial phase φ is 0 [rad].

Further, according to the present embodiment, the fundamental sine waves g _u (t), g _v (t), g _w (t) are combined with the fundamental sine waves g _u (t), g _v (t), g _w (t). ) is used as the signal waves h _ua (t), h _va (t), h _wa (t), and the signal waves h _ua (t), h _va (t), A three-phase PWM drive signal that switches the pulse width at the intersection of h _wa (t) and the carrier wave to form a three-phase pulse width modulated drive voltage is generated, and the basic sine waves g _u (t), g _v ( t), the superimposition rate n _a /m, which is the ratio of the modulation rate n _a of the fifth or higher harmonic to the modulation rate m of g _w (t), is -0.25 or more and -0.05 or less, and 5 The initial phase φ of the harmonics of the next or higher order is greater than or equal to -π/4 [rad] and less than or equal to π/2 [rad]. Because of this, reduction in loss can be expected compared to the case where the fundamental sine waves g _u (t), g _v (t), and g _w (t) are used as signal waves.

Furthermore, according to the present embodiment, since it can be realized by the program of the motor control unit 4, development can be done in a short period of time, development costs can be kept low, and changes can be made flexibly.

Note that the present invention is not limited to the configuration of the above-described embodiments, and elements can be added, deleted, and changed within the scope of the inventive concept.

Furthermore, the present invention can be applied not only to three-phase motors but also to multi-phase motors such as six-phase motors and twelve-phase motors.

[Characteristics evaluation test]
Next, the results of a characteristic evaluation test conducted to determine the configuration of the present invention will be shown.

It is generally thought that as harmonics increase, loss also increases. However, since the material properties of the magnetic material that is the motor core material are nonlinear, it cannot be concluded that loss reduction using harmonics is impossible. Therefore, as a characteristic evaluation test, a ring test and a motor drive test were conducted to evaluate the loss reduction performance of the fifth harmonic superimposed PWM.

In this characteristic evaluation test, the fundamental sine waves g _u (t), g _v (t), g _w (t) are used as signal waves (when the superimposition ratio n/m is 0), and the fundamental sine wave g Tests were conducted using signal waves h _u (t), h _v (t), and h _w (t), which are obtained by superimposing fifth-order harmonics on _u (t), g _v (t), and g _w (t). Is going.

[Characteristics evaluation test 1 (ring test)]
When evaluating loss in a motor, it is difficult to understand basic magnetic properties such as nonlinearity because the motor is a rotating body and the magnetic flux density is distributed. Therefore, we conducted a ring test using a soft magnetic material made of the same material as the motor to evaluate its basic magnetic properties.

FIG. 27 is a schematic configuration diagram of a ring test device 69 related to characteristic evaluation test 1 (ring test). Further, FIG. 28 shows the specifications of the ring sample 61 used in this ring test.

For the iron core material (core material) of ring sample 61, non-oriented electrical steel sheet 35H300 (manufactured by Nippon Steel Corporation) was used. The IGBT inverter 62 is a single-phase Si-IGBT inverter using a power module (PM75RSD060) manufactured by Mitsubishi Electric Corporation. This IGBT inverter 62 is equipped with a Si-IGBT (manufactured by Mitsubishi Electric Corporation, PM75RSD060) as a switching element, and a Si diode (manufactured by Mitsubishi Electric Corporation, RM30TB-H) as a freewheeling diode.

The measurement conditions are as follows: fundamental sine wave frequency f ₁ is 50 [Hz], carrier frequency f _c is 1 [kHz], and input voltage V _dc to IGBT inverter 62 supplied from DC power supply 63 is 15 [V]. The basic sine wave modulation rate m was adjusted so that the sine wave magnetic flux density B _f1 was 1 [T]. It is considered that the constant fundamental sinusoidal magnetic flux density B _f1 in the ring test corresponds to the constant average torque in the motor test.

To measure the primary current _I1 and secondary voltage _V2 , we used a current probe (SS-250) and a voltage probe (SS-320) manufactured by Iwasaki Tsushinki Co., Ltd. and an A/D converter 64 (NI) manufactured by National Instruments. PXI-1031) was used.

For controlling the IGBT inverter 62, a fifth harmonic superimposition PWM method is adopted. That is, a waveform obtained by superimposing the fifth harmonic on the fundamental sine wave g(t) is used as the signal wave h(t), the pulse width is switched at the intersection with the carrier wave, and the IGBT inverter 62 is controlled by the PWM signal.

Next, a method for calculating iron loss in a ring test will be explained. The primary current I ₁ and secondary voltage V ₂ shown in FIG. 27 are measured, and the magnetic field strength H and magnetic flux density B are determined as in the above equations (12) and (13). Further, using the strength H of the magnetic field and the magnetic flux density B, the iron loss P _fe is determined as in the above equation (14). Here, since the magnetic field strength H and magnetic flux density B determined by the above equations (12) and (13) include carrier harmonic components, the magnetization phenomenon becomes complicated. Therefore, in order to clarify the influence of fifth-order harmonic superposition, we investigated the low-order frequency components (f ₁ , f ₃ , f ₅ components: f ₃ is the frequency of the 3rd harmonic of the fundamental sine wave frequency f ₁ , and f ₅ is Consider extracting the fundamental sine wave frequency f (fifth harmonic frequency of ₁ ). The obtained magnetic field strength H and magnetic flux density B are fitted using cftool (approximate curve tool) by numerical calculation software MATLAB (registered trademark) R2019b (The MathWorks, Inc.), and the major loop component H _major , B _major . From these, the major loop iron loss P _major is calculated as in the above equation (15). Further, as in the above equation (16), the difference between the iron loss P _fe and the major loop iron loss P _major is the carrier harmonic iron loss (minor loop iron loss) P _carrier .

<Ring test (superimposition rate characteristics)>
In order to evaluate the characteristics of fifth-order harmonic superposition, the electromagnetic characteristics were measured by changing the superposition ratio n/m in the range of -0.25 to 0.25 in a ring test.

FIG. 29 shows the measurement conditions (overlapping rate characteristics) of the ring test. Note that the initial phase φ of the fifth harmonic was set to 0.

FIG. 30 is a diagram showing the waveform of the signal wave h(t), (a) is a diagram showing the fundamental sine wave g(t), and (b) is a diagram showing the fifth harmonic to the fundamental sine wave g(t). FIG. 3 is a diagram showing a superimposed fifth harmonic superimposed signal. The horizontal axis represents time, and the vertical axis represents signal magnitude. FIG. 30A corresponds to the case where the signal wave is a signal in which the fifth harmonic is not superimposed on the fundamental sine wave g(t). That is, this is a case where the modulation rate n of the fifth harmonic is 0. FIG. 30(b) shows the waveform of the signal wave h(t) when the superimposition ratio n/m is −0.2.

FIG. 31 is a diagram showing the measurement results of the time waveforms of magnetic field strength H and magnetic flux density B by the ring test, and (a) is a diagram showing the measured waveform when the superimposition ratio n/m is -0.2. , (b) is a diagram showing the major loop component when the superposition ratio n/m is -0.2, (c) is a diagram showing the measured waveform when the superposition ratio n/m is -0.1, (d ) is a diagram showing the major loop component when the superimposition ratio n/m is −0.1. Moreover, FIG. 32 is a diagram showing the measurement results of the time waveforms of the magnetic field strength H and the magnetic flux density B by the ring test, and (a) is a diagram showing the measured waveform when the superimposition ratio n/m is 0, (b) is a diagram showing the major loop component when the superimposition ratio n/m is 0, (c) is a diagram showing the measured waveform when the superimposition ratio n/m is 0.2, (d) is a diagram showing the measured waveform when the superimposition ratio n/m is 0.2. FIG. 6 is a diagram showing a major loop component when /m is 0.2.

The horizontal axis shows time, the right vertical axis shows the magnetic field strength H, and the left vertical axis shows the magnetic flux density B.

When the superimposition ratio n/m is 0, this corresponds to the case where the fundamental sine wave g(t) is used as the signal wave h(t) without adding the fifth harmonic. Comparing the cases where the superposition ratio n/m is -0.2, -0.1, 0.2 and the case where the superposition ratio n/m is 0, the time waveforms of the magnetic field strength H and magnetic flux density B are It can be seen that the shapes are different, and changes appear in the magnetization phenomenon due to fifth-order harmonic superposition.

FIG. 33 is a diagram showing the measurement results of the BH curve of magnetic field strength H and magnetic flux density B by the ring test, and (a) is the measurement result when the superimposition ratio n/m is -0.2 (solid line) (b) is a diagram showing the measurement results (solid line) and the major loop component (dashed line) when the superposition ratio n/m is -0.1, (c) is the superposition ratio A diagram showing the measurement results (solid line) and the major loop component (broken line) when n/m is 0, and (d) shows the measurement results (solid line) and the major loop component ( (dashed line). This figure shows the measurement data shown in FIGS. 31 and 32, with the horizontal axis representing the magnetic field strength H and the vertical axis representing the magnetic flux density B. Note that the graph indicated by the broken line indicates the major loop components H _major and B _major calculated by the above method.

FIG. 34 is a diagram showing measurement results of the fundamental wave current I _f1 and the fundamental sine wave modulation factor m by the ring test. The horizontal axis is the superposition ratio n/m, the vertical axis on the left is the fundamental wave current I _f1 , and the vertical axis on the right is the fundamental sine wave modulation rate m.

The fundamental wave current I _f1 is an excitation current component for obtaining the fundamental wave magnetic flux density B _f1 , and increases slightly when the superimposition ratio n/m is greater than 0, and greatly decreases when the superposition ratio n/m is smaller than 0. A trend was observed. Under a constant DC voltage V _dc to the IGBT inverter 62, the basic sine wave modulation rate m is adjusted so that the basic sine wave magnetic flux density B _f1 becomes 1 [T], and the superimposition rate n/m is greater than 0. Then, the basic sine wave modulation rate m increases, and when the superimposition rate n/m becomes smaller than 0, the basic sine wave modulation rate m decreases. In other words, it can be said that due to the superposition of the fifth harmonic, the fundamental wave current I _f1 decreases while the fundamental sinusoidal magnetic flux density B _f1 is constant.

FIG. 35 is a diagram showing measurement results of iron loss P _fe , major loop iron loss P _major , and carrier harmonic iron loss (minor loop iron loss) P _carrier by a ring test, and (a) is a diagram showing iron loss. , (b) are diagrams showing the rate of change in iron loss when the superimposition ratio n/m is 0 as a reference.

FIG. 35(b) shows the results when the superposition ratio n/m is changed based on the iron loss P _fe , the major loop iron loss P _major , and the carrier harmonic iron loss P _carrier when the superposition ratio n/m is 0. It shows the rate of change of iron loss P _fe , major loop iron loss P _major , and carrier harmonic iron loss P _carrier .

In a range where the superimposition ratio n/m is larger than 0, the iron loss P _fe increases, and in a range where the superimposition ratio n/m is smaller than 0, the iron loss P _fe decreases. Further, when the superimposition ratio n/m was −0.2, the iron loss P _fe showed the minimum value, and the iron loss reduction rate was 3.3%. The major loop iron loss P _major decreases when the overlap ratio n/m is −0.15, −0.1, and −0.05. When the superimposition ratio n/m is −0.1, the major loop iron loss P _major is at its minimum, and the reduction rate is 2.3%. The carrier harmonic iron loss P _carrier increases when the superposition ratio n/m is greater than 0, and decreases when the superposition ratio n/m is smaller than 0. When the superimposition ratio n/m is −0.25, it is minimum, and the iron loss reduction rate of the carrier harmonic iron loss P _carrier is 17.8%.

Compared to the case where the 5th harmonic is not superimposed (when the superimposition ratio n/m is 0), the iron loss P _fe is reduced when the superposition ratio n/m is in the range of -0.25 or more and -0.05 or less. are doing.

When extrapolating the graph of the rate of change of iron loss P _fe shown in FIG. It can be said that the iron loss P _fe is reduced even when the superimposition ratio n/m is −0.3 or more, compared to the case where n/m is 0. Further, in the graph of the rate of change of the iron loss P _fe shown in FIG. 35(b), the iron loss P _fe decreases even when the superimposition ratio n/m is in the range of 0 and −0.05.

<Ring test (phase angle characteristics of 5th harmonic)>
In order to evaluate the characteristics of the initial phase φ of the fifth harmonic, the initial phase φ was varied in the range of −3π/4 [rad] to 3π/4 [rad] in a ring test, and the electromagnetic characteristics were measured.

FIG. 36 is a diagram showing the measurement conditions (phase angle characteristics of the fifth harmonic) of the ring test. Note that the superimposition ratio n/m was fixed at -0.2. Further, in this measurement, the DC voltage V _dc to the IGBT inverter 62 was adjusted so that the fundamental sinusoidal magnetic flux density B _f1 of the ring sample 61 was 1 [T].

FIG. 37 is a diagram showing the waveform of the signal wave h(t), (a) is a diagram showing the fundamental sine wave g(t), and (b) is a diagram showing the fundamental sine wave g(t) with an initial phase φ of π. FIG. 3 is a diagram showing a fifth-order harmonic superimposed signal on which a fifth-order harmonic of /4 [rad] is superimposed. The horizontal axis represents time, and the vertical axis represents signal magnitude. FIG. 37A corresponds to the case where the modulation rate n of the fifth harmonic is 0 and the signal wave is a signal in which the fifth harmonic is not superimposed on the fundamental sine wave g(t). FIG. 37(b) shows the waveform of the signal wave h(t) when the superimposition ratio n/m is −0.2.

FIG. 38 is a diagram showing the measurement results when the initial phase φ of the fifth harmonic was changed by the ring test, (a) is a diagram showing the fundamental wave current I _f1 , and (b) is a diagram showing the iron loss P _fe FIG. The horizontal axis is the initial phase φ, the vertical axis is the fundamental wave current I _f1 in FIG. 38(a), and the iron loss P _fe in FIG. 38(b). In addition, the horizontal broken line in the figure indicates the fundamental wave current I _f1 when the modulation rate n of the fifth harmonic is 0, that is, when the fundamental sine wave g(t) is used as the signal wave h(t), It shows the measured value of iron loss _Pfe .

Compared to the horizontal broken line, the iron loss P _fe is reduced in the range where the initial phase φ is -π/4 [rad] or more and π/2 [rad] or less, and when the initial phase φ is π/4 [rad] ] is the minimum value. In addition, by organizing this measurement data and connecting the plots where the initial phase φ is π/4 [rad] and π/2 [rad] with a straight line, the iron loss P _fe is smaller than the plot where the initial phase φ is 0. It can be seen that the range is such that the initial phase φ is greater than or equal to π/4 [rad] and less than or equal to 5π/16 [rad]. From this figure, it can be confirmed that even when the initial phase φ is in the range of 3π/8 [rad] or more and π/4 [rad] or less, the iron loss P _fe is smaller than when the initial phase φ is 0. . In other words, when the initial phase φ of the fifth harmonic is set in the range of 3π/8 [rad] or more and 5π/16 [rad] or less, the iron loss P _fe becomes smaller than when the initial phase φ is 0. .

From the above, if you want to take advantage of the effect of reducing iron loss P _fe , you can set the initial phase φ to a range of -π/4 [rad] or more and π/2 [rad] or less, or set the initial phase φ to a range of -π/4 [rad] or more and π/2 [rad] or more, or Moreover, it can be said that it has been confirmed that it is effective to set it to a range of 5π/16 [rad] or less, or to set it to π/4 [rad].

<Ring test (carrier frequency characteristics)>
In order to evaluate the characteristics of the carrier frequency f _c of the fifth harmonic, the carrier frequency f _c was varied in the range of 1 [kHz] to 20 [kHz] in a ring test, and the electromagnetic characteristics were measured.

FIG. 39 is a diagram showing the measurement conditions (carrier frequency characteristics) of the ring test, in which (a) shows the basic measurement conditions and (b) shows the superimposition conditions of the fifth harmonic. As shown in FIG. 39(b), the superimposition ratio n/m is fixed at -0.2, the initial phase φ of the fifth harmonic is 0 in case X, and the initial phase φ is π/4 [rad ]. Further, in this measurement, the DC voltage V _dc to the IGBT inverter 62 was adjusted so that the fundamental sinusoidal magnetic flux density B _f1 of the ring sample 61 was 1 [T].

FIG. 40 is a diagram showing the measurement results when changing the carrier frequency f _c by the ring test, (a) is a diagram showing the fundamental wave current I _f1 , (b) is a diagram showing the iron loss P _fe . . The horizontal axis is the carrier frequency f _c , the vertical axis is the fundamental wave current I _f1 in FIG. 40(a), and the iron loss P _fe in FIG. 40(b). The diamond plots show the measurement results for case X, and the triangle plots show the measurement results for case Y. Moreover, the circle plot shows the measurement results when the fifth harmonic is not superimposed (when the superimposition ratio n/m is 0).

As shown in FIG. 40(a), the fundamental _wave current I _f1 in case It is the smallest in the range. The fundamental wave current I _f1 in case Y (superimposition rate n/m is set to -0.2 and initial phase φ is set to π/4 [rad]) is 5 in the range where the carrier frequency f _c is 10 [kHz] or more. This is smaller than when the harmonics are not superimposed.

As shown in FIG. 40(b), the iron loss P _fe in case Y (the superimposition ratio n/m is set to -0.2 and the initial phase φ is set to π/4 [rad]) _is is the smallest in the range. _The iron loss P _fe in case It is smaller than when it is not superimposed.

[Characteristics evaluation test 2 (motor test)]
Next, the measurement results of the motor test are shown. A motor drive test was conducted under conditions in which the superimposition ratio n/m was 0 or less, and the loss characteristics of the motor drive test device 89 were evaluated.

FIG. 41 is a schematic configuration diagram of a motor testing device 89 related to characteristic evaluation test 2 (motor test). In this motor test, an IGBT inverter 71 is used to drive an embedded permanent magnet synchronous motor (IPMSM) 73.

FIG. 42 is a diagram showing a test motor (embedded structure permanent magnet synchronous motor 73) for the motor test, (a) is a schematic cross-sectional view of the motor, and (b) is a diagram showing the specifications of the motor.

This embedded structure permanent magnet synchronous motor 73 is composed of a rotor and a stator, and the core material of the rotor and stator is non-oriented electrical steel plate 35H300 (manufactured by Nippon Steel Corporation), similar to the ring sample 61 of the ring test. be. The permanent magnet is a bonded magnet (manufactured by Aichi Steel Corporation, S5P-12ME). The IGBT inverter 71 is a three-phase Si-IGBT inverter that uses a power module (PM75RSD060) manufactured by Mitsubishi Electric Co., Ltd. as a switching element, and a Si diode (RM30TB-H, manufactured by Mitsubishi Electric Co., Ltd.) as a freewheeling diode. ing. Power measurement and waveform observation are performed using Yokogawa PX-8000 (power meter 72).

The measurement conditions were that the rotational speed ω was 750 [rpm], the average torque T was 0.611 [Nm], the carrier frequency _fc was 1 [kHz], and the input voltage V _dc to the IGBT inverter 71 was 50 [V]. . The fundamental sine wave modulation factor m was adjusted by performing feedback control so that the rotational speed ω and the torque T were constant.

Next, a loss calculation method of the motor testing device 89 will be explained. Input power P _in to the IGBT inverter 71, input power P _u , P _v , P _w of each phase of the motor, effective input current value I _{u_rms} , I _{v_rms} , I _{w_rms} , input current I _u of each phase of the motor , I _v , and I _w are measured and used to calculate the loss. As shown in the above equation (17), the overall loss P _total of the test device 89 is composed of the inverter loss P _inv , the copper loss P _Cu , and the motor core loss/mechanical loss P _core&mech . The inverter loss P _inv is the input power P _in to the IGBT inverter 71 , the input power P _u , P _v , P _w of each phase of the motor, the loss P _{w of the power measuring device 72 . m} is calculated as in the above equation (18). Loss P of power meter 72 _{w. m} is determined by the shunt resistance R _shunt (=0.1 [Ω]), the resistance R _cable (=0.012 [Ω]) of the connection cable, and the effective input current values of each phase of the motor I _{u_rms} , I _{v_rms} , I _{w_rms} It is calculated as in the above equation (19). Copper loss P _Cu is calculated using the winding resistance R (=0.5 [Ω]) and the effective input current values of each phase of the motor I _{u_rms} , I _{v_rms} , and I _{w_rms} as shown in the above formula (20). The motor core loss/mechanical loss P _core&mech is calculated as shown in the above formula (21) using the input power P _u , P _v , P _w of each phase of the motor, the copper loss P _Cu , and the mechanical output ωT. The results are the average value of seven measurements, and the error is expressed as standard deviation. In addition, since it is difficult to directly measure both motor core loss P _core and mechanical loss P _mech , in this measurement, mechanical loss P _mech and motor core loss P _core are not classified, and motor core loss/mechanical loss P _core&mech The measurement results were obtained as follows.

<Motor test (superimposition rate characteristics)>
In order to evaluate the characteristics of 5th harmonic superposition, the loss characteristics were measured by varying the superposition ratio n/m in the range of -0.25 to 0 in a motor test.

FIG. 43 shows the measurement conditions (superimposition rate characteristics) of the motor test. Note that the initial phase φ of the fifth harmonic was set to 0.

FIG. 44 is a diagram showing the waveforms of three-phase signal waves h _u (t), h _v (t), and h _w (t), and (a) is a diagram showing the waveforms of the three-phase signal waves h _u (t), h v (t), and g _v ( t), g _w (t), (b) is a 5th harmonic superimposed signal in which the 5th harmonic is superimposed on the fundamental sine waves g _u (t), g _v (t), and g _w (t). FIG. The horizontal axis represents time, and the vertical axis represents signal magnitude. FIG. 44(a) corresponds to the case where the signal wave is a signal in which the fifth harmonic is not superimposed on the fundamental sine waves g _u (t), g _v (t), and g _w (t). That is, this is a case where the modulation rate n of the fifth harmonic is 0. FIG. 44(b) shows the waveforms of three-phase signal waves h _u (t), h _v (t), and h _w (t) when the superimposition ratio n/m is −0.1.

FIG. 45 is a diagram showing the measurement results of the fundamental wave current I _{f1_rms} and the fundamental sine wave modulation factor m in the motor test.

This figure shows the fifth order of the fundamental sine wave modulation rate m obtained by feedback control so that the rotational speed ω (=750 [rpm]) and the average torque T (=0.611 [Nm]) are constant. It shows the superimposition rate characteristics of harmonics. The fundamental sine wave modulation rate m is -0.25, -0.15, -0. It is decreasing by 1. Furthermore, the phase average fundamental wave current I _{f1_rms} obtained from the observed waveform also has a superimposition ratio n/m of −0. 25, -0.15, and -0.1. This is the same phenomenon as the measurement result of the ring test (see FIG. 34).

FIG. 46 shows the measurement results of the overall loss P _total in the motor test. The horizontal axis is the superimposition ratio n/m, and the vertical axis is the overall loss P _total . In addition, the horizontal broken line in the figure is the measurement of the total loss P _total when the modulation rate n of the fifth harmonic is 0, that is, when the fundamental sine wave g(t) is used as the signal wave h(t). It shows the value.

The overall loss P _total is greater when the superposition rate n/m is -0.25, -0.15, and -0.1 compared to the case where the fifth harmonic is not superimposed (when the superposition ratio n/m is 0). is decreasing. Further, when the superimposition ratio n/m is -0.15, it shows the minimum value, and the loss reduction rate is 1.5% compared to the case where the fifth harmonic is not superimposed. Also, when the superimposition ratio n/m is −0.10, the loss reduction rate is 1.4%, and the loss is greatly reduced.

In this way, it was confirmed that the loss reduction effect is high when the superimposition ratio n/m is -0.15 and -0.10. Further, from this measurement result, it is easily inferred that the loss is reduced even when the superimposition ratio n/m is -0.05, which is between 0 and -0.10.

From the above, in order to reduce loss, the superimposition ratio n/m is set to a range of -0.25 or more and -0.05 or less, or set to a range of -0.15 or more and -0.10 or less. This is effective.

In addition, as shown below, in this motor test (superposition ratio characteristics), when the superposition ratio n/m is -0.15, -0.10, the inverter loss P _inv and copper loss P which constitute the overall loss P _total are All losses including _Cu , motor core loss and mechanical loss P _{core & mech} were reduced compared to the case where the fifth harmonic was not superimposed. In _other _words , the _{superimposition} ratio is Fifth-order harmonic superposition with n/m of −0.15 and −0.10 may be effective.

FIG. 47 shows the measurement results of motor core loss and mechanical loss P _core&mech by motor test. The horizontal broken line in the figure is the measurement of motor core loss/mechanical loss P _core&mech when the fundamental sine wave g(t) is used as the signal wave h(t) (when the modulation rate n of the fifth harmonic is 0). It shows the value.

Motor core loss/mechanical loss P _core&mech decreases when the superposition ratio n/m is -0.15 and -0.1 compared to when the fifth harmonic is not superimposed (when the superposition ratio n/m is 0) are doing. Furthermore, when the superimposition ratio n/m is -0.10, it shows the minimum value, and the loss reduction rate is 2.5% compared to the case where the fifth harmonic is not superimposed. Here, it is assumed that the mechanical loss P _mech is constant since the rotational speed is constant. In this case, it is considered from the above ring test that the reason for the decrease in motor core loss P _core is a decrease in major loop iron loss P _major and carrier harmonic iron loss P _carrier . In particular, the trend is very similar to the trend of the major loop iron loss P _major in the ring test, including the case where the superimposition ratio n/m is −0.25 (see FIG. 35).

FIG. 48 shows the measurement results of copper loss P _Cu and fundamental wave current copper loss P _{Cu_If1} in the motor test. The horizontal broken line in the figure indicates the measured value when the fundamental sine wave g(t) is used as the signal wave h(t) (when the modulation rate n of the fifth harmonic is 0).

The fundamental wave current copper loss P _{Cu_If1} is calculated similarly to the above formula (20) using the fundamental wave current I _{f1_rms} in FIG. 45. The copper loss P _Cu decreases when the superposition ratio n/m is -0.15 and -0.1 compared to when the fifth harmonic is not superimposed (when the superposition ratio n/m is 0). . Further, when the superimposition ratio n/m is -0.10, it shows the minimum value, and the loss reduction rate is 0.4% compared to the case where the fifth harmonic is not superimposed. As the absolute value of the superposition ratio n/m increases, the harmonic copper loss (=P _Cu -P _{Cu_If1} ) increases, while the fundamental current copper loss P _{Cu_If1} decreases. From this, it can be seen that the copper loss P _Cu is due to a decrease in the fundamental wave current I _{f1_rms} .

FIG. 49 shows the measurement results of the inverter loss P _inv in the motor test.

The inverter loss P _inv is greater when the superposition ratio n/m is -0.25, -0.15, and -0.1 compared to the case where the fifth harmonic is not superimposed (when the superposition ratio n/m is 0). is decreasing. Further, when the superimposition ratio n/m is -0.25, it shows the minimum value, and the loss reduction rate is 5.1% compared to the case where the fifth harmonic is not superimposed.

<Motor test (5th harmonic phase angle characteristics)>
In order to evaluate the characteristics of the initial phase φ of the fifth harmonic, loss characteristics were measured by setting the initial phase φ to 0 [rad] and π/4 [rad] in a motor test.

FIG. 50 shows the measurement conditions for the motor test (phase angle characteristics of the fifth harmonic). Note that the superimposition ratio n/m was set to -0.1.

FIG. 51 is a diagram showing the waveforms of three-phase signal waves h _u (t), h _v (t), h _w (t), and (a) is a diagram showing the waveforms of the three-phase signal waves h _u (t), h v (t), and g _v ( (b) _{is a diagram showing fundamental sine waves g u} ₍ t), g _v (t), g _w (t) with an initial phase φ of π/4 [rad]. FIG. 3 is a diagram showing a fifth-order harmonic superimposed signal on which harmonics are superimposed. The horizontal axis represents time, and the vertical axis represents signal magnitude. FIG. 51(a) corresponds to the case where the signal wave is a signal on which the fifth harmonic is not superimposed on the fundamental sine waves g _u (t), g _v (t), g _w (t) (fifth harmonic (When the harmonic modulation rate n is 0). FIG. 51(b) shows three-phase signal waves h _u (t), h when a fifth-order harmonic with a superposition ratio n/m of −0.1 and an initial phase φ of π/4 [rad] is superimposed. These are the waveforms of _v (t) and h _w (t).

FIG. 52 shows the measurement results of the fundamental sine wave modulation factor m and the fundamental wave current I _{f1_rms} in the motor test. On the horizontal axis, the left side is when the superposition rate n/m is 0, the center is when the superposition rate n/m is -0.1 and the initial phase φ of the fifth harmonic is 0 [rad], and the right side is the superposition rate n/m. The case where m is −0.1 and the initial phase φ is π/4 [rad] is shown.

FIG. 53 is a diagram showing the measurement results of the overall loss P _total in the motor test. On the horizontal axis, the left side is when the superposition rate n/m is 0, the center is when the superposition rate n/m is -0.1 and the initial phase φ of the fifth harmonic is 0 [rad], and the right side is the superposition rate n/m. The case where m is −0.1 and the initial phase φ is π/4 [rad] is shown, and the vertical axis is the overall loss P _total . In addition, the horizontal broken line in the figure is the measured value of the overall loss P _total when the fundamental sine wave g(t) is used as the signal wave h(t) (when the modulation rate n of the fifth harmonic is 0). It shows.

As described later, when comparing the test results when the initial phase φ of the fifth harmonic is 0 [rad] and π/4 [rad] under the condition that the superposition ratio n/m is -0.1, the following is obtained. It became so.

The reduction rate of the motor core loss/mechanical loss P _{core &mech} in the case where the fifth harmonic is not superimposed showed a larger value when the initial phase φ was π/4 [rad] than when it was 0 [rad]. The reduction rate of copper loss P _Cu showed a slightly larger value when the initial phase φ was 0 [rad] than when the initial phase φ was 0 [rad]. The reduction rate of the inverter loss P _inv was larger when the initial phase φ was 0 [rad] than when the initial phase φ was π/4 [rad].

Based on this result, the following configuration of the invention may be considered. Under low speed/high torque conditions, the copper loss P _Cu and inverter loss P _inv are assumed to be large, and under high speed/low torque conditions, the motor core loss/mechanical loss P _core&mech is assumed to be large.

In this case, a predetermined threshold value such as a predetermined rotation speed threshold value is set in advance, and if a command value such as a rotation speed command that is a motor control command input from the outside is smaller than this predetermined threshold value, the 5th harmonic The initial phase φ of the wave is set to 0 [rad], and when the command value is greater than or equal to a predetermined threshold, the initial phase φ is switched to π/4 [rad], superimposed on the fundamental sine wave g(t), and the signal wave h(t ) may be generated.

In addition, a predetermined threshold value such as a predetermined torque threshold value is set in advance, and if a command value such as a torque command that is a motor control command input from the outside is greater than or equal to this predetermined threshold value, the fifth harmonic The phase φ is set to 0 [rad], and if the command value is smaller than a predetermined threshold, the initial phase φ is switched to π/4 [rad] and superimposed on the fundamental sine wave g(t) to generate the signal wave h(t). You may also do so.

In addition, instead of the input command value, the rotational speed and torque of the operating motor are detected, and when the detected rotational speed and torque reach a predetermined threshold, the initial phase φ of the fifth harmonic is switched. You may also do so.

Furthermore, the initial phase φ may be switched smoothly in a linear or curved manner instead of discontinuously when the predetermined threshold is reached. In addition, two predetermined thresholds, a smaller predetermined threshold and a larger predetermined threshold, are prepared, and when the detected rotational speed or torque reaches the smaller predetermined threshold, the initial phase φ of the fifth harmonic is switched. Then, when the detected rotational speed or torque reaches the larger predetermined threshold and the initial phase φ is switched, the initial phase φ is then set to the smaller predetermined threshold. The predetermined threshold value may be set with hysteresis.

In this way, the initial phase φ of the fifth harmonic to be superimposed on the fundamental sine wave g(t) may be switched based on input values such as a command value to the motor or a detected value of the motor during operation. .

In addition, a predetermined threshold value such as a predetermined rotation speed threshold value is set in advance, and when a command value such as a rotation speed command input from the outside is smaller than this predetermined threshold value, the initial phase φ is changed from 0 [rad]. The signal is superimposed on the fundamental sine wave g(t) so that the initial phase φ becomes π/4 [rad] when the command value is greater than or equal to a predetermined threshold. A wave h(t) may be generated. In this case, π/4 [rad] is the maximum initial phase φ added to the fifth harmonic. The method for continuously changing the initial phase φ may be linear or curved.

In addition, a predetermined threshold value such as a predetermined torque threshold value is set in advance, and when a command value such as a torque command input from the outside is greater than or equal to this predetermined threshold value, the initial phase φ is changed from π/4 [rad] to 0. [rad], and when the command value is smaller than a predetermined threshold value, the signal wave h( t) may be generated.

Alternatively, instead of the input command value, the rotational speed and torque of the operating motor may be detected, and the initial phase φ of the fifth harmonic may be changed based on the detected rotational speed and torque. good.

In this way, the initial phase φ of the fifth harmonic superimposed on the fundamental sine wave g(t) is continuously changed based on the input values such as the command value to the motor and the detected value of the motor during operation. You may also do so.

That is, the initial phase φ of the fifth harmonic to be superimposed is not a fixed value, but may be changed. Furthermore, the state of change in the initial phase φ of the fifth harmonic may be switched based on a predetermined threshold value.

FIG. 54 shows the measurement results of motor core loss and mechanical loss P _core&mech by motor test. The horizontal broken line in the figure is the measurement of motor core loss/mechanical loss P _core&mech when the fundamental sine wave g(t) is used as the signal wave h(t) (when the modulation rate n of the fifth harmonic is 0). It shows the value.

The motor core loss/mechanical loss P _core&mech is reduced when the superposition ratio n/m is −0.1 compared to the case where the fifth harmonic is not superimposed (the case where the superposition ratio n/m is 0). Furthermore, under the condition that the superimposition ratio n/m is -0.1, the motor core loss/mechanical loss P _core&mech is reduced when the initial phase φ of the fifth harmonic is π/4 [rad] than when it is 0 [rad]. . The loss reduction rate when the initial phase φ is π/4 [rad] is 3.4% compared to the case where the fifth harmonic is not superimposed.

In "high speed/low torque conditions where motor core loss/mechanical loss Pcore _&mech is expected to be large", it is considered more effective to set the initial phase φ of the fifth harmonic to π/4 [rad].

FIG. 55 shows the measurement results of copper loss P _Cu and fundamental wave current copper loss P _{Cu_If1} in the motor test. The horizontal broken line in the figure indicates the measured value of the copper loss P _Cu when the modulation rate n of the fifth harmonic is 0.

The copper loss P _Cu is reduced when the superposition ratio n/m is -0.1 compared to the case where the fifth harmonic is not superimposed (when the superposition ratio n/m is 0). Furthermore, under the condition where the superimposition ratio n/m is -0.1, the copper loss P _Cu of 0 [rad] is reduced compared to the case where the initial phase φ of the fifth harmonic is π/4 [rad]. . The loss reduction rate when the initial phase φ is 0 [rad] is 0.4% compared to the case where the fifth harmonic is not superimposed.

The fundamental wave current copper loss P _{Cu_If1} is reduced when the superimposition ratio n/m is -0.1, compared to the case where the fifth harmonic is not superimposed (when the superposition ratio n/m is 0). In addition, under the condition where the superimposition ratio n/m is -0.1, the fundamental wave current copper loss P _{Cu_If1} of π/4 [rad] is reduced compared to the case where the initial phase φ of the fifth harmonic is 0 [rad]. are doing.

FIG. 56 shows the measurement results of the inverter loss P _inv in the motor test. The horizontal broken line in the figure indicates the measured value of the inverter loss P _inv when the modulation rate n of the fifth harmonic is 0.

The inverter loss P _inv is reduced when the superposition ratio n/m is -0.1 compared to the case where the fifth harmonic is not superimposed (the case where the superposition ratio n/m is 0). Furthermore, under the condition where the superposition ratio n/m is -0.1, the inverter loss P _inv of 0 [rad] is reduced compared to the case where the initial phase φ of the fifth harmonic is π/4 [rad]. . The loss reduction rate when the initial phase φ is 0 [rad] is 2.2% compared to the case where the fifth harmonic is not superimposed.

DESCRIPTION OF SYMBOLS 1... Motor drive system, 2... Three-phase inverter part, 3... Boost chopper part, 4... Motor control part, 5... Permanent magnet synchronous motor, 6... Stator coil, 7... Rotor, _S1 , _S2 , _S3 , S ₄ , S ₅ , S ₆ ... switching element, D ₁ , D ₂ , D ₃ , D ₄ , D ₅ , D ₆ ... free-wheeling diode, 8 ₁ , 8 ₂ , 8 ₃ , 8 ₄ , 8 ₅ , 8 ₆ ...Switching element input section, 9...Current sensor, 10...Position sensor, 11...Battery, 12...Inductor, 13...Capacitor, Sc...Switching element for chopper section, 15...Diode, 40...CPU, 41...ROM, 42... RAM, 43... Signal wave generation section, 44... Carrier wave generation section, 45... PWM drive signal generation section, 46... PWM drive signal output section, 47... Boost chopper control signal output section, 48... Rotor detection position reception section, 49 ...Motor input current value acceptance unit, 50...Command value acceptance unit, 60...5th harmonic superposition PWM controller, 61...Ring sample, 62...IGBT inverter, 63...DC power supply, 64...A/D converter, 69... Ring test device, 70... Fifth harmonic superposition PWM controller, 71... IGBT inverter, 72... Power measuring instrument, 73... Embedded structure permanent magnet synchronous motor (IPMSM), 74... Encoder, 75, 76... Torque meter, 77 ...Power analyzer, 78...Load, 79...BLDC motor, 80...Rectifier, 81...MCU & MOSFET, 89...Motor test equipment, 109...Ring test equipment, 100...Fifth harmonic superposition PWM controller, 101...Ring sample, 102... IGBT inverter, 103... DC power supply, 104... A/D converter, 108 ₁ , 108 ₂ , 108 ₃ , 108 ₄ ... Switching element input section, 289... Motor testing device, 270... Fifth harmonic superposition PWM controller, 271 ... IGBT inverter, 272 ... Power measuring device, 273 ... Embedded structure permanent magnet synchronous motor, 274 ... Encoder, 275 ... Torque meter, 276 ... Torque meter, 277 ... Power analyzer, 278 ... Load, 279 ... BLDC motor, 280 ... Rectifier, 281...MCU&MOSFET

Claims

A motor drive system that drives a motor with an inverter,
In the switching operation of the semiconductor in the inverter at the intersection of the signal wave h(t) obtained by superimposing harmonics on the fundamental sine wave g(t) and the carrier wave,
The fundamental sine wave g(t) is g(t)=m・sin(2πf 1 t),
The signal wave h(t) is h(t)=m・sin(2πf 1 t)+n・sin(2π・5f 1 t+φ),
The upper limit of n/m is the value of n/m at which the fundamental wave current starts to decrease due to changes in magnetic properties,
The lower limit of n/m is a value of n/m at which the increase due to harmonic components is greater than the reduction effect of the fundamental wave current,
A motor drive system that operates with
The motor drive system according to claim 1, characterized in that it operates with n/m greater than -0.3 and less than 0.
When the phase angle (2πf 1 t) of the fundamental sine wave g(t) is π/2 radians, the numerical value m·sin(π/2) of the fundamental sine wave g(t) is the signal wave h(t )=m・sin(π/2)+n・sin(5・π/2+φ) or more, the motor drive according to claim 1 or 2, characterized in that the motor operates at an initial phase φ of the harmonics that is greater than or equal to a numerical value of system.
The maximum phase angle of the initial phase φ of the harmonics is the value of the initial phase φ at which the fundamental current starts to decrease due to a change in the magnetic characteristics;
a value of the initial phase φ at which the minimum phase angle of the initial phase φ is increased by a harmonic component more than the fundamental wave current reduction effect;
3. The motor drive system according to claim 1, wherein the motor drive system operates as follows.
A motor drive system that drives a motor with an inverter,
In the switching operation of the semiconductor in the inverter at the intersection of the signal wave h(t) obtained by superimposing harmonics on the fundamental sine wave g(t) and the carrier wave,
The fundamental sine wave g(t) is g(t)=m・sin(2πf 1 t),
The signal wave h(t) is h(t)=m・sin(2πf 1 t)+n・sin(2π・af 1 t+φ),
a is an integer of 5 or more,
The upper limit of n/m is the value of n/m at which the fundamental wave current starts to decrease due to changes in magnetic properties,
The lower limit of n/m is a value of n/m at which the increase due to harmonic components is greater than the reduction effect of the fundamental wave current,
A motor drive system that operates with
The motor drive system according to claim 5, characterized in that it operates with n/m greater than -0.3 and less than 0.
When the phase angle (2πf 1 t) of the fundamental sine wave g(t) is π/2 radians, the numerical value m·sin(π/2) of the fundamental sine wave g(t) is the signal wave h(t )=m·sin(π/2)+n·sin(a·π/2+φ) The motor drive according to claim 5 or 6, characterized in that the motor operates at an initial phase φ of the harmonic that is equal to or greater than a value of n·sin(a·π/2+φ). system.
The maximum phase angle of the initial phase φ of the harmonics is the value of the initial phase φ at which the fundamental current starts to decrease due to a change in the magnetic characteristics;
a value of the initial phase φ at which the minimum phase angle of the initial phase φ is increased by a harmonic component more than the fundamental wave current reduction effect;
The motor drive system according to claim 5 or 6, characterized in that the motor drive system operates as follows.
A motor drive system comprising: an inverter section that outputs a pulse width modulated drive voltage for driving a synchronous motor; and a motor control section that controls the inverter section to form the pulse width modulation drive voltage.
The motor control section is
The pulse width modulation drive voltage is changed by switching the pulse width at the intersection of a carrier wave and a signal wave in which a fifth harmonic of the fundamental sine wave is superimposed on a fundamental sine wave that defines the fundamental frequency of the pulse width modulation drive voltage. The PWM drive signal is configured to generate a PWM drive signal and supply the PWM drive signal to a switching element input section of the inverter section,
When the frequency of the fundamental sine wave is the fundamental sine wave frequency f1 , the signal wave h(t) is set as h(t)=m・sin(2πf 1 t)+n・sin(2π・5f 1 t+φ), A ring formed of the iron core material of the synchronous motor, the upper and lower limits of a superimposition rate n/m being a ratio of the modulation rate n of the fifth harmonic of the initial phase φ to the modulation rate m of the fundamental sine wave. A pulse width modulated voltage using the signal wave h(t) is applied to the primary coil wound around the sample while changing the superposition ratio n/m, and the primary current flowing through the primary coil and the ring sample are It is determined by a ring test that measures the secondary voltage generated in the wound secondary coil.
The upper limit of the superimposition rate n/m is
The superimposition ratio n/m is the upper limit of a range in which the fundamental wave current, which is a component of the fundamental sine wave frequency f1 of the primary current, is lower than when the modulation rate n of the fifth harmonic is zero, or The superimposition ratio n/m is determined as the upper limit of a range in which the predetermined loss calculated based on the primary current and the secondary voltage is lower than when the modulation ratio n of the fifth harmonic is zero. ,
The lower limit of the superimposition rate n/m is
Compared to the amount of reduction in the fundamental wave current based on the case where the modulation rate n of the fifth harmonic is zero, the amount of reduction in the primary current is based on the case where the modulation rate n of the fifth harmonic is zero. The superimposition ratio n/m, which is the lower limit of the range below which the amount of increase in the harmonic current, which is a harmonic component of the fundamental sine wave frequency f1 , or the modulation ratio n of the fifth harmonic is zero. A motor drive system characterized in that the superimposition ratio n/m is determined as the lower limit of a range below which the predetermined loss falls.
Maximum and minimum phase angles of the initial phase φ of the fifth harmonic are determined by the ring test in which the initial phase φ is varied and the pulse width modulated voltage is applied to the primary coil. and
The maximum phase angle is
The initial phase φ is the maximum value of the range in which the fundamental wave current is lower than when the modulation rate n of the fifth harmonic is zero, or the initial phase φ is lower than when the modulation rate n of the fifth harmonic is zero. The initial phase φ is determined as the maximum value in the range below a predetermined loss,
The minimum phase angle is
Compared to the amount of reduction in the fundamental wave current based on the case where the modulation rate n of the fifth-order harmonic is zero, the harmonic current is reduced based on the case where the modulation rate n of the fifth-order harmonic is zero. Determine the initial phase φ as the minimum value in the range below which the increase amount is lower, or the initial phase φ as the minimum value in the range in which the predetermined loss is lower than when the modulation rate n of the fifth harmonic is zero. The motor drive system according to claim 9, characterized in that:
A motor drive system comprising: an inverter section that outputs a pulse width modulated drive voltage for driving a synchronous motor; and a motor control section that controls the inverter section to form the pulse width modulation drive voltage.
The motor control section is
The pulse width modulation drive voltage is changed by switching the pulse width at the intersection of a carrier wave and a signal wave in which a fifth harmonic of the fundamental sine wave is superimposed on a fundamental sine wave that defines the fundamental frequency of the pulse width modulation drive voltage. The PWM drive signal is configured to generate a PWM drive signal and supply the PWM drive signal to a switching element input section of the inverter section,
When the frequency of the fundamental sine wave is the fundamental sine wave frequency f1 , the signal wave h(t) is set as h(t)=m・sin(2πf 1 t)+n・sin(2π・5f 1 t+φ), The upper and lower limits of the superimposition rate n/m, which is the ratio of the modulation rate n of the fifth harmonic of the initial phase φ to the modulation rate m of the fundamental sine wave, are determined by the The synchronous motor is rotationally driven by changing the superimposition ratio n/m by a pulse width modulation drive voltage, and the input power to the inverter section, the input power of each phase of the synchronous motor, or the input current of each phase of the synchronous motor is It is determined by a motor test that measures at least one of the following:
The upper limit of the superimposition rate n/m is
The superimposition ratio n/m is the upper limit of a range in which the fundamental wave current, which is a component of the fundamental sine wave frequency f1 calculated from the input current, is lower than when the modulation rate n of the fifth harmonic is zero. or, the predetermined loss calculated based on at least one of the input power to the inverter section, the input power to each phase of the synchronous motor, or the input current to each phase of the synchronous motor is the fifth harmonic. The superimposition rate n/m is determined as the upper limit of the range below the modulation rate n when it is zero,
The lower limit of the superimposition ratio n/m is:
Compared to the amount of reduction in the fundamental wave current based on the case where the modulation rate n of the fifth harmonic is zero, the input current is reduced based on the case where the modulation rate n of the fifth harmonic is zero. The superimposition ratio n/m, which is the lower limit of the range below which the amount of increase in the harmonic current, which is a harmonic component of the fundamental sine wave frequency f1 , or the modulation ratio n of the fifth harmonic is zero. A motor drive system characterized in that the superimposition ratio n/m is determined as the lower limit of a range below which the predetermined loss falls.
The maximum phase angle and the minimum phase angle of the initial phase φ of the fifth harmonic are determined by the motor test in which the initial phase φ is varied and the synchronous motor is rotationally driven by the pulse width modulated drive voltage. It is
The maximum phase angle is
The initial phase φ is the maximum value of the range in which the fundamental wave current is lower than when the modulation rate n of the fifth harmonic is zero, or the initial phase φ is lower than when the modulation rate n of the fifth harmonic is zero. The initial phase φ is determined as the maximum value in the range below a predetermined loss,
The minimum phase angle is
Compared to the amount of reduction in the fundamental wave current based on the case where the modulation rate n of the fifth-order harmonic is zero, the harmonic current is reduced based on the case where the modulation rate n of the fifth-order harmonic is zero. Determine the initial phase φ as the minimum value in the range below which the increase amount is lower, or the initial phase φ as the minimum value in the range in which the predetermined loss is lower than when the modulation rate n of the fifth harmonic is zero. The motor drive system according to claim 11, characterized in that: