WO2016067665A1

WO2016067665A1 - Inverter control device and inverter compressor

Info

Publication number: WO2016067665A1
Application number: PCT/JP2015/063909
Authority: WO
Inventors: アンヴァンホ; 恵子多田; 哲英横山; 佐竹　彰; 瀧口　隆一; 将吾諸江
Original assignee: 三菱電機株式会社
Priority date: 2014-10-30
Filing date: 2015-05-14
Publication date: 2016-05-06
Also published as: CN106537761B; CN106537761A

Abstract

The purpose of the present invention is to realize an inverter control device capable of determining the operation mode of a machine to be driven such as a compressor. The inverter control device (17) according to the present invention comprises: an operation mode determining unit (7) which estimates power or shaft torque of an electric motor (1) operated in an operation mode having a variation component frequency of power or shaft torque in power variation in one rotation in mechanical angle, from current in a power line through which control voltage is applied to the electric motor (1), and determines the operation mode of the electric motor (1) on the basis of a frequency component relevant to the variation component frequency of the estimated power or the estimated torque that has been estimated; a rotation speed command value generating unit (13) which generates a rotation speed command (ωe*) according to the operation mode determined by the operation mode determining unit (7); and an inverter output voltage control unit (12) which generates a voltage command (voltage command vector (Vuvw*)) for the control voltage applied to the electric motor (1), on the basis of the value of the rotation speed command (ωe*).

Description

Inverter control device and inverter compressor

The present invention relates to an inverter control device for controlling an electric motor and an inverter compressor using the same.

In recent years, since the Kyoto Protocol, which includes greenhouse gas emission reduction targets for the prevention of global warming, was adopted in 2005, CO2 emission reduction regulations have been strengthened in each country, and energy-saving regulation standards for air conditioners have become strict. It has become. In Europe, unlike domestic indicators, the efficiency at the rated time is not evaluated, but the one that evaluates the energy saving performance based on the annual power consumption is the index. Since the method of calculating the efficiency of each heating is adopted, the efficiency improvement at low load is a new technical issue.

A rotary compressor has two compression chambers, and has a structure that switches between a single operation in which only one of the two compression chambers is compressed and a parallel operation in which both are compressed, thereby improving the compression efficiency. A compressor has been developed. In this case, it is necessary to discriminate switching between parallel operation and single operation and to perform appropriate control according to the difference in the refrigerant circulation amount.

For example, Patent Document 1 discloses a two-cylinder rolling piston type rotary compressor, in which a half-capacity operation is automatically selected when the load is small. A configuration for halving the circulation flow rate is disclosed. In this configuration, since the motor can be operated without reducing the rotation speed of the electric motor that rotates the piston of the rotary compressor, the compressor efficiency can be improved.

JP 2009-203861 A (0049 to 0052 stages, FIG. 1)

In the two-cylinder rotary compressor of Patent Document 1, in order to prevent the ability of the two-cylinder rotary compressor from changing when switching between the single operation and the parallel operation, that is, to not change the refrigerant circulation flow rate. It is necessary to match the output frequency of the inverter that controls the motor to the operation mode. Therefore, in order to stably control each operation mode, it is necessary to correctly grasp the current operation mode in the compressor. However, the method is not disclosed in the rotary compressor of Patent Document 1 for determining the operation mode, and when the operation mode is switched, the optimum operation is not performed and the operation is performed with incorrect control. Could occur.

The present invention has been made to solve the above-described problems, and an object of the present invention is to realize an inverter control device that can determine an operation mode of a machine to be driven such as a compressor.

An inverter control device according to the present invention is an inverter control device that controls an electric motor that drives a machine to be driven based on an AC voltage converted by an inverter, and the electric motor has electric power in a power fluctuation during one rotation of a mechanical angle or The motor power or shaft torque is estimated from the current of the power line that applies the control voltage to the motor and is operated in the operation mode having the fluctuation component frequency of the shaft torque, and related to the estimated power or the fluctuation component frequency of the estimated torque. Based on the frequency component, an operation mode determination unit that determines the operation mode of the motor, a rotation speed command value generation unit that generates a rotation speed command according to the operation mode determined by the operation mode determination unit, a rotation speed command And an inverter output voltage control unit that generates a voltage command of a control voltage to be applied to the electric motor based on the value. .

The inverter control device according to the present invention estimates the electric power or shaft torque of the electric motor, and calculates a frequency component related to the estimated component of fluctuation of the estimated electric power or estimated torque. Can be determined.

It is a block diagram which shows the inverter control apparatus and inverter compressor by Embodiment 1 of this invention. It is a figure which shows the coordinate converter of FIG. It is a figure which shows the electric power estimation part of FIG. It is a figure which shows the electric power pulsation extraction part of FIG. It is a figure which shows the discrimination | determination signal production | generation part of FIG. It is a figure which shows the inverter output voltage control part of FIG. It is a figure which shows the axial torque fluctuation | variation in one rotation in the electric motor of FIG. It is a figure which shows the extraction result of the electric power pulsation extraction part of FIG. 1 in case one compression part operate | moves. It is a figure which shows the extraction result of the electric power pulsation extraction part of FIG. 1 in case two compression parts operate | move. It is a figure which shows the extraction result of the electric power pulsation extraction part of FIG. 1 when three compression parts operate | move. It is a figure which shows the extraction result of the electric power pulsation extraction part of FIG. 1 when four compression parts operate | move. It is a figure which shows the analysis result of the pulsation component of the electric power of a scroll machine and a single machine by the electric power pulsation extraction part of FIG. It is a figure which shows the other discrimination | determination signal production | generation part of FIG. It is a figure which shows the hardware constitutions of the inverter control apparatus and inverter compressor by Embodiment 1 of this invention. It is a block diagram which shows the inverter control apparatus and inverter compressor by Embodiment 2 of this invention. It is a figure which shows the electric power estimation part and output torque estimation part of FIG. It is a figure which shows the electric power pulsation extraction part of FIG. It is a figure which shows the inverter output voltage control part of FIG. It is a figure which shows the torque control part of FIG. It is a figure explaining the rotational speed change by torque pulsation. It is a figure explaining compensation of torque pulsation by Embodiment 2 of the present invention. It is a figure explaining the current vector by Embodiment 2 of this invention. It is a figure which shows the vibration suppression result by Embodiment 2 of this invention. It is a figure which shows the compressor by Embodiment 3 of this invention. It is a schematic cross-sectional view in the 1st compression part of FIG. It is a schematic cross-sectional view in the 2nd compression part of FIG. It is a block diagram which shows the inverter control apparatus and inverter compressor by Embodiment 3 of this invention. It is a figure which shows the driving | operation mode switching detection part and torque pulsation extraction part of FIG. It is a figure which shows the example of control of the compressor by Embodiment 3 of this invention. It is a figure which shows the example of control of the compressor by Embodiment 3 of this invention. It is a block diagram which shows the inverter control apparatus and inverter compressor by Embodiment 4 of this invention. FIG. 32 is a Bode diagram of a leading phase filter in the phase adjustment unit of FIG. 31. It is a figure which shows the vibration suppression result in the case of no lead phase filter by Embodiment 4 of this invention. It is a figure which shows the vibration suppression result in the case of the advance phase filter presence by Embodiment 4 of this invention. It is a block diagram which shows the inverter control apparatus and inverter compressor by Embodiment 5 of this invention. It is a figure which shows the torque pulsation extraction part of FIG. It is a figure which shows the cosine wave generation part and sine wave generation part of FIG. It is a figure which shows the flow chat of the learning part of FIG. It is a figure which shows the waveform of instruction | command (omega) e of the rotational speed of an electric motor. It is a block diagram which shows the inverter control apparatus and inverter compressor by Embodiment 6 of this invention. It is a figure which shows the torque command value switching part of FIG. It is a figure explaining the torque learning part of FIG. It is a figure which shows the structure of the electric motor phase estimation part of FIG. It is a figure which shows the angle memory | storage part and estimated torque memory | storage part of FIG. FIG. 43 is a diagram in which data is recorded in an angle storage unit and an estimated torque storage unit in FIG. 42. It is a figure which shows the flow chat of the learning algorithm process part of FIG. It is a figure which shows the flow chat of the recording mode execution process of FIG. It is a figure which shows the flow chat of the output mode execution process of FIG. It is a figure which shows the waveform of the load torque and output torque of an electric motor before learning by the learning algorithm process part of FIG. It is a figure which shows the waveform of the load torque and output torque of an electric motor after the learning by the learning algorithm process part of FIG. It is a figure which shows the rotational speed of the electric motor before and behind learning by the learning algorithm process part of FIG. It is a figure which shows the FFT analysis result of the rotational speed of the electric motor before the learning by the learning algorithm process part of FIG. It is a figure which shows the FFT analysis result of the rotational speed of the electric motor after the learning by the learning algorithm process part of FIG.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing an inverter control device and an inverter compressor according to Embodiment 1 of the present invention. FIG. 2 is a diagram showing the coordinate converter of FIG. 1, and FIG. 3 is a diagram showing the power estimation unit of FIG. 4 is a diagram illustrating the power pulsation extracting unit of FIG. 1, and FIG. 5 is a diagram illustrating the discrimination signal generating unit of FIG. FIG. 6 is a diagram illustrating the inverter output voltage control unit of FIG. The inverter compressor 100 includes an inverter control device 17 and a compressor 80. The inverter control device 17 controls the electric motor 1 of the compressor 80. In FIG. 1, an inverter control device 17 outputs and rotates a three-phase control voltage to an electric motor 1 used for a compressor 80 that circulates refrigerant in N (N is an integer of 1 or more) compressors. . The electric motor 1 is a synchronous or brushless electric motor. The inverter 16 outputs a three-phase AC PWM (Pulse Width Modulation) voltage with a predetermined frequency and amplitude to the electric motor 1 through the three-phase power line using the DC power source 15.

Current sensors

2, 3, and 4 detect a current flowing through electric motor 1, that is, a current flowing through a three-phase power line. The current sensor 2 detects a current flowing through the u-phase three-phase power supply line. The

current sensors

3 and 4 detect currents flowing through the v-phase and w-phase three-phase power supply lines, respectively. As appropriate, the u-phase three-phase power supply line, the v-phase three-phase power supply line, and the w-phase three-phase power supply line are referred to as a three-phase power supply line u, a three-phase power supply line v, and a three-phase power supply line w, respectively. The current detection unit 5 calculates a three-phase current from the outputs of the

current sensors

2, 3 and 4.

The inverter control device 17 includes a coordinate converter 6, an operation mode determination unit 7, a rotation speed command value generation unit 13, an inverter output voltage control unit 12, an inverter gate signal generation unit 14, an inverter 16, and a current detection unit 5. . The coordinate converter 6 converts coordinates from the AC three-phase current detected by the current detector 5 to a two-phase DC current. The rotational speed command value generation unit 13 generates a rotational speed command value of the electric motor 1. The inverter output voltage control unit 12 generates a control voltage for driving the electric motor 1 at the commanded rotation speed. The operation mode determination unit 7 generates the determination signal hnt based on the dq axis current vector Idq and the dq axis voltage command vector Vdq *. The operation mode determination unit 7 includes a power estimation unit 8, a power pulsation extraction unit 9, and a determination signal generation unit 10. The power estimation unit 8 converts the dq-axis voltage command vector Vdq * output from the inverter output voltage control unit 12 and the dq-axis current vector Idq converted based on the coordinate converter 6 from the current value calculated by the current detection unit 5. Based on this, the electric power of the motor 1 is estimated, and the estimated electric power P ^ is calculated. The power pulsation extraction unit 9 extracts the amplitude || xkf || of the pulsation component (ripple component) of the power detected from the estimated power P ^ at a predetermined frequency. k is an integer from 1 to N, and the amplitude of the pulsating component exists from || x1f || to || xNf ||. The discrimination signal generator 10 generates a discrimination signal hnt that discriminates the operation mode in which the compressor 80 is currently operating based on the dominant pulsation component in the extracted pulsation component. The determination signal generation unit 10 determines a selection frequency that is a frequency of a dominant pulsation component in the extracted pulsation component, and determines an operation mode associated with the selection frequency. The inverter control device 17 having the above configuration can control the input of the electric motor 1 so as to set a predetermined rotational speed in the operation mode of the compressor 80.

The inverter control device 17 will be described in detail. 1 includes

current sensors

2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3 and 3. The detected current vector Idt (see FIG. 2) whose components are the three-phase currents iu, iv, and iw is measured from the four signals. However, it is good also as a structure which provides a current sensor (for example, current sensor 2, 3) in two phases among the three phases of the electric motor 1, and obtains the three-phase electric current value of the electric motor 1 by the following formula | equation (1).

As shown in FIG. 2, the coordinate converter 6 includes a coordinate converter 85 and a coordinate converter 86. The coordinate converter 85 converts the three-phase currents iu, iv, iw, which are components of the detected current vector Idt, into two-phase currents iα, iβ, which are fixed two-phase currents. The coordinate converter 86 converts the two-phase currents iα and iβ into a dq-axis current vector Idq based on the estimated rotation angle θ ^ e of the rotor of the electric motor 1. The components of the dq axis current vector Idq are a d axis current id and a q axis current iq. The coordinate converter 85 converts the three-phase currents iu, iv, iw of the detected current vector Idt into the two-phase currents iα, iβ based on the equation (2). Expression (2) is a conversion expression from three-phase coordinates (u, v, w) to fixed two-phase coordinates (α, β). In addition, the coordinate converter 86 converts the two-phase currents iα and iβ into a d-axis current id and a q-axis current iq based on Expression (3). Expression (3) is a conversion expression from fixed two-phase coordinates (α, β) to rotating two-phase coordinates (d, q). The estimated rotation angle θ ^ e is calculated from a magnetic flux estimation unit 18 and an integrator 19 (see FIG. 6) described later.

As shown in FIG. 3, the power estimation unit 8 is based on the current detection unit 5 and the coordinate converter 6 and converts the d-axis current id of the dq-axis current vector of the electric motor 1, the q-axis current iq, and the inverter output voltage. The estimated power P ^ of the electric motor 1 is calculated from the d-axis command voltage vd * and the q-axis command voltage vq * of the dq-axis voltage command vector Vdq * output by the control unit 12. That is, based on the d-axis current id, the q-axis current iq, the d-axis command voltage vd *, and the q-axis command voltage vq * of the electric motor 1, the estimated power P ^ is calculated as in Expression (4).

As shown in FIG. 3, the circuit that realizes the calculation of Expression (4) includes two

multipliers

26 a and 26 b and an adder 27. The inputs of the multiplier 26a are the d-axis command voltage vd * and the d-axis current id. The inputs of the multiplier 26b are the q-axis command voltage vq * and the q-axis current iq. The adder 27 adds the outputs of the

multipliers

26a and 26b.

As shown in FIG. 4, the power pulsation extraction unit 9 generates the amplitude of the pulsation component of the power corresponding to the number of compression units in order to determine the operation mode of the compressor 80, that is, the number of compression units being operated. To do. An example in which the compressor 80 includes N compression units will be described. FIG. 7 is a diagram showing shaft torque fluctuation in one rotation in the electric motor of FIG. FIG. 7 shows shaft torque fluctuations when k of the N compression units are operating. The horizontal axis is the rotation angle, and the vertical axis is the load torque. FIG. 7 shows the case where the value of k is 1, 2, 3, 4. Note that k is an integer greater than 0 and less than or equal to N. The torque characteristic 91a is a case where k is 1 and one compression unit is operating. The torque characteristic 91b is when k is 2 and when two compression units are operating. The torque characteristic 91c is a case where k is 3 and three compression units are operating. The torque characteristic 91d is a case where k is 4 and four compression units are operating. As shown in FIG. 7, at one rotation of the electric motor 1 (0 to 360 degrees of mechanical angle), power pulsation due to fluctuations in the pressure inside the compressor 80 or fluctuations in the shaft torque of the electric motor 1 occurs.

The power pulsation extraction unit 9 uses the fact that the k compression units in operation are out of timing for compressing the refrigerant, so that the electric power pulsation extraction unit 9 performs one rotation of the motor 1 (mechanical angle 0 to 360 degrees). The power pulsation caused by the fluctuation of the pressure inside the compressor 80 or the fluctuation of the shaft torque of the electric motor 1 is extracted. In the motor 1 of the compressor 80 that compresses the refrigerant by shifting the phase by (360 / k) degrees in k compression sections in one rotation of the motor 1, the mechanical rotation frequency of the motor 1 (hereinafter referred to as frequency f1f) is used as electric power. And pulsation having a frequency (fkf = k × f1f) times as high as that of k. A frequency k times the machine rotation frequency is defined as a frequency fkf. As a result, in the case of the compressor 80 that compresses the refrigerant with k compression units in a certain time, a pulsating component having a frequency fkf is generated in the electric power of the electric motor 1. By determining the frequency fkf of the pulsating component, the operation mode determination unit 7 can determine the operation mode of the compressor 80, that is, the number of compression units being operated. Specifically, the operation mode discriminating unit 7 extracts the magnitude of the pulsating component corresponding to the possible frequencies from the electric power of the motor 1, that is, N frequencies (f1f,..., Fkf,..., FNf). Based on the dominant pulsating component of these pulsating components, it is possible to determine the operation mode of the compressor 80, that is, the number of operating compression units.

As shown in FIG. 4, the power pulsation extraction unit 9 includes N power pulsation component extractors. Each power pulsation component extractor calculates the magnitude of a pulsation component having a specific frequency. In FIG. 4, three power

pulsation component extractors

81a, 81b, 81c are specifically shown. The code | symbol of an electric power pulsation component extractor uses 81 generally, and 81a, 81b, 81c is used when distinguishing. The power pulsation component extractor 81 performs a well-known discrete Fourier transform (DFT) operation on the estimated power P ^ estimated by the power estimation unit 8 at a predetermined frequency fkf, and the frequency component xkf Find the amplitude of. The discrete Fourier transform was performed as shown in Equation (5).

Here, M is an integer of 1 or more in the number of intervals of one period T. Further, expression (5) becomes as follows by expansion of complex number e ^ (jωktm) of expression (5).

Since the amplitudes of xkA and xkB on the right side of Equation (6) coincide with the integration of one period T, Equation (7) is obtained. Further, the amplitude of xkf is as shown in Expression (8).

Based on the equations (5) to (8), in the electric motor 1 of the compressor 80 having N compression units, there is a possibility that pulsation components having different frequencies are generated in the electric power depending on the number of compression units in operation. Therefore, the power pulsation extraction unit 9 includes N power pulsation component extractors. The power pulsation component extractor 81a calculates the amplitude of the pulsation component having the frequency f1f. The power pulsation component extractor 81b calculates the amplitude of the pulsation component of the frequency fkf, and the power pulsation component extractor 81c calculates the amplitude of the pulsation component of the frequency Nf. The configuration of the power pulsation component extractor 81 will be described using the power pulsation component extractor 81b as an example.

The rotation angle θk based on the predetermined frequency fkf at time t is 2 * π * fkf * t. The power pulsation component extractor 81b includes a sine wave generator 29b and a cosine wave generator 28b that generate a sine component (sin component) and a cosine component (cos component) at the rotation angle θk, and the respective components and the estimated power P ^. Multipliers 30c and 30d,

integrators

31c and 31d for integrating the values obtained from the multipliers as shown in equation (7), and integrated || xkA || and || xkB || ) Is provided with an amplitude calculator 32b for calculating the magnitude of the vibration component of the predetermined frequency, that is, the amplitude || xkf ||. The integrator 31c calculates || xkA ||, and the integrator 31d calculates || xkB ||.

Calculating other frequency components based on the same configuration. The power pulsation component extractor 81a includes a cosine wave generator 28a, a sine wave generator 29a,

multipliers

30a and 30b,

integrators

31a and 31b, and an amplitude calculator 32a. The power pulsation component extractor 81c includes a cosine wave generator 28c, a sine wave generator 29c,

multipliers

30e and 30f,

integrators

31e and 31f, and an amplitude calculator 32c. The power pulsating component extractor 81a calculates the amplitude || x1f || of the pulsating component of the predetermined frequency f1f at time t, and the power pulsating component extractor 81c calculates the amplitude of the pulsating component of the predetermined frequency Nf || xNf at time t. || is calculated.

8 to 11 show the amplitudes (magnitudes) of the pulsating components of the four frequencies f1f, f2f, f3f, and f4f when N = 4. 8 to 11, pulsation components of frequencies f1f, f2f, f3f, and f4f are pulsation components of

frequency orders

1f, 2f, 3f, and 4f, respectively. FIG. 8 is a diagram illustrating an extraction result of the power pulsation extraction unit of FIG. 1 when one compression unit is operated, and FIG. 9 is a diagram of the power pulsation extraction unit of FIG. 1 when two compression units are operated. It is a figure which shows an extraction result. FIG. 10 is a diagram illustrating an extraction result of the power pulsation extraction unit of FIG. 1 when three compression units are operated, and FIG. 11 is a diagram of the power pulsation extraction unit of FIG. 1 when four compression units are operated. It is a figure which shows an extraction result. 8 to 11, the horizontal axis is the frequency order, and the vertical axis is the power. As shown in FIG. 8, in the operation mode of the compressor 80 in which only one compressor (k = 1) circulates the refrigerant, the power pulsation component having the frequency f1f is dominant. In this case, the frequency f1f is set as the selected frequency. As shown in FIG. 9, in the operation mode of the compressor 80 in which the refrigerant is circulated by the two compression units (k = 2), the power pulsation component having the frequency f2f is large. In this case, the frequency f2f is set as the selected frequency. As shown in FIG. 10, in the operation mode of the compressor 80 that circulates three compressors (k = 3) refrigerant, the power pulsation component having the frequency f3f is dominant. In this case, the frequency f3f is selected. As shown in FIG. 11, in the operation mode of the compressor 80 that compresses all the compressors (k = 4) refrigerant, the power pulsation component having the frequency f4f is large. In this case, the frequency f4f is selected.

FIG. 5 shows a determination signal generator 10 that determines the operation mode based on the magnitude (amplitude) of the pulsation component. The determination signal generation unit 10 includes a maximum amplitude detection unit 33 and a determination unit 34. The maximum amplitude detection unit 33 compares the amplitudes || x1f || to || xNf || of the pulsation components output to the power pulsation extraction unit 9 with each other, detects the frequency having the maximum amplitude, Output as U. That is, the frequency having the maximum amplitude is obtained by the function argmax shown in Expression (9). In FIG. 5, only three amplitudes || x1f ||, || xkf ||, and || xNf || are shown.

Based on the value of the function value U, the determination unit 34 that determines based on the following language expression determination rule determines the operation mode in which the compressor 80 operates, and generates a determination signal hnt.
When U = mf, hnt = m. In this case, it determines with the operation mode which circulates a refrigerant | coolant with m compression parts. Here, m is an integer from 1 to N. Three cases are specifically shown below.
When U = 1f, hnt = 1. In this case, it determines with the operation mode which circulates a refrigerant | coolant by one compression part.
When U = kf, hnt = k. In this case, it is determined as an operation mode in which the refrigerant is circulated by the k compression units.
When U = Nf, hnt = N. In this case, it determines with the operation mode which circulates a refrigerant | coolant in all the compression parts.

As shown in FIG. 1, the rotation speed command value generation unit 13 determines each operation mode based on the determination signal hnt output from the determination signal generation unit 10 and the optimum operation speed command value for each operation mode stored in advance. Rotational speed command ωe * according to is generated. For example, when the electric motor 1 is driven so that the refrigerant circulation flow rate of the compressor 80 becomes constant, the rotation speed command ωe * is set as follows.
When hnt = m, ωe * = ωe0 / m. Here, m is an integer from 1 to N. ωe0 is a reference rotational speed command value, and is a rotational speed command value when one compression unit is performing compression operation. Three cases are specifically shown below.
When hnt = 1, ωe * = ωe0.
When hnt = k, ωe * = ωe0 / k.
When hnt = N, ωe * = ωe0 / N.

As shown in FIG. 6, the inverter output voltage control unit 12 includes a magnetic flux estimation unit 18, an integrator 19, a subtracter 20, a speed control unit 21, a current control unit 22, and a voltage coordinate converter 23. The inverter output voltage control unit 12 performs speed control for feeding back the estimated rotational speed ω ^ e based on the estimation of the magnetic flux of the electric motor 1. In the present embodiment, the electric motor 1 of the compressor 80 is in the refrigerant, and it is difficult to attach an encoder or the like that detects the position of the rotor of the electric motor 1. Therefore, the inverter output voltage control unit 12 employs the configuration shown in FIG.

Depending on the operation mode in which the compressor 80 is operated, the amplitude and fluctuation frequency of the electric power of the electric motor 1 are different even with the same differential pressure. Based on the above determination, it is preferable to generate a rotation speed command of the electric motor 1 in accordance with the operation mode so that the electric power is suitable for the differential pressure.

The speed controller 21 receives the value of the rotational speed command ωe * generated by the rotational speed command value generator 13 and the speed difference Δω between the estimated rotational speed ω ^ e estimated by the magnetic flux estimator 18 and inputs the estimated rotational speed ω ^. A dq-axis current command vector Idq * is generated and output so that e matches the value of the rotational speed command ωe *. Here, the speed difference Δω is ωe * −ω ^ e. The value of the rotational speed command ωe * is appropriately expressed as a rotational speed command value ωe *. The components of the dq-axis current command vector Idq * are a d-axis current command id * and a q-axis current command iq *.

The current control unit 22 converts the dq axis current command vector Idq * input from the speed control unit 21 and the detected current vector Idt of the motor 1 detected by the current detection unit 5 into a dq axis current vector Idq by the coordinate converter 6. Based on this, the dq-axis voltage command vector Vdq * is output so that the dq-axis current vector Idq matches the dq-axis current command vector Idq *.

The voltage coordinate converter 23 reverses the input and output of the d-axis command voltage vd * and the q-axis command voltage vq * in the calculated dq-axis voltage command vector Vdq * from the relationship of the formulas (2) and (3). Based on the inverse transformation, the three-phase command voltages vu *, vv *, and vw * in the voltage command vector Vuvw * are converted. As described above, the inverter output voltage control unit 12 generates the voltage command vector Vuvw * so that the current of the motor 1 matches the dq-axis current command vector Idq *. The inverter gate signal generation unit 14 (see FIG. 1) outputs a gate signal for controlling on / off (ON / OFF) of each switching element of the inverter 16 using the voltage command vector Vuvw *.

6 uses the dq-axis current vector Idq and the dq-axis voltage command vector Vdq * to calculate the estimated rotational speed ω ^ e. First, the dq-axis magnetic flux vector in the state space expression is expressed by the differential equation of Expression (10) using the dq-axis current vector Idq and the dq-axis voltage command vector Vdq *.

However, Φ ^ ds is a magnetic flux of d-axis electron reaction, and Φ ^ qs is a magnetic flux of q-axis electron reaction. Φ ^ dr is the d-axis rotor magnetic flux. R is the resistance of the armature, and Ld and Lq are the d-axis inductance and the q-axis inductance in the armature, respectively. h11, h12, h21, h22, h31, and h32 are set feedback gains. Note that the d-axis estimated current i ^ d and the q-axis estimated current i ^ q based on the magnetic flux vectors (Φ ^ ds, Φ ^ qs, Φ ^ dr) can be expressed as in Expression (11). Further, the estimated rotational speed ω ^ e of the electric motor 1 can be calculated based on the equation (12). However, i ^ d and i ^ q indicate the d-axis estimated current and the q-axis estimated current in the magnetic flux estimating unit 18.

However, kap is an acceleration estimated proportional gain, and ωapi is an acceleration estimated integral gain.

As a result, the estimated rotation angle θ ^ e of the rotor of the electric motor 1 is calculated as shown in Expression (13) based on the integrator 19 in FIG.

The inverter control device 17 according to the first embodiment is configured as described above so that the dq-axis current converted from the detected current vector Idt of the motor 1 of the compressor 80 that fluctuates according to the pressure fluctuation or the shaft torque fluctuation inside the compressor 80. Based on the vector Idq and the dq axis voltage command vector Vdq * that drives the rotation of the motor 1, the output power of the motor 1 is estimated, and the amplitude of each frequency component calculated in real time from the estimated power P ^ || x1f By comparing ||, || xkf ||, and || xNf ||, the current compression operation mode in the compressor 80 can be determined.

Further, the inverter control device 17 of the first embodiment does not use a position sensor that detects the position of the rotor of the electric motor 1, and the magnetic flux Φ ^ ds of the d-axis electron reaction that is the estimated magnetic flux, q-axis electric Based on the provision of the inverter output voltage control unit 12 that estimates the rotational speed of the electric motor 1 based on the magnetic flux Φ ^ qs of the child reaction and the d-axis rotor magnetic flux Φ ^ dr and calculates the estimated estimated rotational speed ω ^ e. The speed control of the electric motor 1 can be performed stably. Further, since the position sensor for detecting the position of the rotor of the electric motor 1 is not used, the cost of the position sensor of the compressor 80 controlled based on the inverter control device 17 and the inverter 16 can be reduced.

Furthermore, the determination method for determining the compression operation mode of the compressor 80 performed by the inverter control device 17 of the first embodiment is not only suitable for the compressor 80 having one or more compression units, but also has different pressure fluctuation patterns. The compression operation mode of the compressor 80 can be determined. Even in the case of the compressor 80 having different pressure fluctuation patterns, the compression operation mode can be determined based on the amplitudes || x1f ||, || xkf ||, || xNf || Needless to say, the compression operation mode can be determined even in the case of the compressor 80 having one pressure motion pattern.

In this case, the determination rule in the determination unit 34 of the determination signal generation unit 10 is changed from the rule described above to another. For example, an example will be described in which it is determined whether the operation mode is a compression operation mode with a scroll compressor having a pair of spiral bodies or a compression operation mode with a single compressor having one compression unit. The frequency of the dominant power pulsation component in both the scroll compressor having a pair of spiral bodies and the single compressor having one compression unit is f1f, and the discrimination rule in the discrimination unit 34 described so far is It is difficult to discriminate whether it is a scroll machine or a single machine based on it.

The discrimination rules for discriminating whether a scroll machine or a single machine is shown below. FIG. 12 is a diagram illustrating an analysis result of the power pulsation component of the scroll machine and the single machine by the power pulsation extraction unit 9 of FIG. The horizontal axis in FIG. 12 is the frequency order, and the vertical axis is power. As shown in FIG. 12, the ratio s (|| x0f |) of the strength (amplitude || x0f ||) of the 0f component (average value) of power and the strength (amplitude || x1f ||) of the 1f component | / || x1f ||), the ratio s of the scroll machine is larger than the ratio s of the single machine. Therefore, this feature is used to configure the discrimination signal generator 10 shown in FIG. FIG. 13 is a diagram illustrating another determination signal generation unit in FIG. 1. The discrimination signal generation unit 10 in FIG. 13 includes a ratio calculation unit 90 and a discrimination unit 34. The ratio calculation unit 90 calculates a ratio s (|| x0f || / | x1f ||) between the strength of the 0f component (average value) and the strength of the 1f component of the electric power. The discrimination rules in the discrimination unit 34 are set as follows.

When s> 5, it is determined as a scroll machine, and the value of the determination signal hnt is set to 10, for example.
When s <3, it is determined as a single machine, and the value of the determination signal hnt is set to 1, for example.

Therefore, the inverter control device 17 including the determination signal generation unit 10 shown in FIG. 13 can determine the operation of the scroll compressor and the operation of the single rotary compressor (single machine). Even when there are a plurality of models of the compressor 80, the rotation speed command value generation unit 13 is configured to determine the optimum operation speed for each operation mode stored in advance by configuring the determination signal generation unit 10 that determines the model of the compressor 80. Based on the command and the determination signal hnt output from the determination signal generation unit 10, it is possible to generate a rotation speed command value ωe * according to the model of the compressor 80.

Although the example in which the closed loop control is performed in the first embodiment of the present invention has been described, the inverter control device 17 sets the operation mode determination unit 7 even when performing the open loop control such as the magnetic flux vector control and the V / f control. Therefore, the compression operation mode of the compressor 80 can be determined.

Moreover, the method for estimating the rotational speed of the electric motor 1 described above is an example of a method for estimating the rotational speed of the electric motor 1 without using a position sensor, and a rotational speed estimation method other than the above is used. Also good.

Since the inverter compressor 100 according to the first embodiment includes the inverter control device 17 according to the first embodiment, the operation mode of the compressor 80 can be determined. Moreover, the inverter compressor 100 of Embodiment 1 automatically detects the switching of the operation mode in the compressor 80 in which the number of compression units that compress the refrigerant changes, and changes the rotational speed ωe of the electric motor 1 for each mode. By controlling to do so, the refrigerant circulation flow rate can be made constant.

As described above, the inverter control device 17 according to the first embodiment is an inverter control device that controls the electric motor 1 that drives the drive target machine (compressor 80) based on the AC voltage converted by the inverter 16. The electric motor 1 is operated in an operation mode having a fluctuation component frequency of electric power or shaft torque in electric power fluctuation during one rotation of the mechanical angle, and a power supply line (three-phase power supply lines u, v, w) for applying a control voltage to the electric motor 1 An operation mode discriminating unit for estimating the operation mode of the electric motor 1 based on the frequency component related to the fluctuation component frequency of the estimated estimated power P ^ or the estimated torque τ ^. 7, a rotation speed command value generation unit 13 that generates a rotation speed command ωe * according to the operation mode determined by the operation mode determination unit 7, and the value of the rotation speed command ωe *. Wherein the inverter output voltage control unit 12 that generates a voltage command of the control voltage applied to the motor 1 (* voltage command vector Vuvw), further comprising a Te. Based on this configuration, the inverter control device 17 of the first embodiment estimates the electric power or shaft torque of the electric motor 1 and calculates a frequency component related to the fluctuation component frequency of the estimated electric power P ^ or the estimated torque τ ^. Therefore, the operation mode of the drive target machine (compressor 80) can be determined.

Inverter compressor 100 of the first embodiment includes a plurality of compression units, an electric motor 1 that drives all of the compression units with one rotating shaft, and an inverter control that controls the electric motor 1 based on an alternating voltage converted by an inverter 16. A device 17 is provided. The inverter control device 17 of the inverter compressor 100 is an inverter control device that controls the electric motor 1 that drives the drive target machine (compressor 80) based on the AC voltage converted by the inverter 16, and the electric motor 1 is a machine. The motor 1 is operated from an electric current of a power supply line (three-phase power supply lines u, v, w) that is operated in an operation mode having a fluctuation component frequency of electric power or shaft torque in the electric power fluctuation during one rotation of the angle. An operation mode discriminating unit 7 that discriminates an operation mode of the electric motor 1 based on a frequency component related to a fluctuation component frequency of the estimated estimated power P ^ or estimated torque τ ^. A rotation speed command value generation unit 13 that generates a rotation speed command ωe * according to the operation mode determined by the determination unit 7, and an electric power based on the value of the rotation speed command ωe *. An inverter output voltage control unit 12 that generates a voltage command of the control voltage applied to the machine 1 (* voltage command vector Vuvw), characterized by comprising a. Based on this configuration, inverter compressor 100 of the first embodiment estimates the power or shaft torque of motor 1 and calculates a frequency component related to the estimated component P ^ or the fluctuation component frequency of estimated torque τ ^. Therefore, the operation mode of the drive target machine (compressor 80) can be determined.

The fluctuation component frequency is a frequency of one or more fluctuation components of the shaft torque or power resulting from the periodic fluctuation of the machine load during one rotation of the machine angle of the electric motor 1. Includes the frequency of the pulsating component.

The rotational speed command value generation unit 13, the coordinate converter 6, the operation mode determination unit 7, the inverter output voltage control unit 12, and the inverter gate signal generation unit 14 in the functional block diagram of FIG. This is realized by the processor 301 that executes the program stored in the memory. FIG. 14 is a diagram illustrating a hardware configuration of the inverter control device and the inverter compressor according to the present embodiment. The inverter control device 17 includes a processor 301, a storage device 302, a current detection unit 5, and an inverter 16. In the storage device 302, the function program in each embodiment is stored in advance. The processor 301 executes a function program stored in the storage device 302. The processor 301 communicates with the host controller 300. In addition, a plurality of processors 301 and a plurality of storage devices 302 may cooperate to execute the above function program. Furthermore, the processor 301 and the storage device 302 may not only execute the functions of the present invention but also simultaneously perform other functions such as control of a DC power supply and communication with a remote controller.

Embodiment 2. FIG.
FIG. 15 is a block diagram showing an inverter control device and an inverter compressor according to the second embodiment of the present invention. 16 is a diagram illustrating the power estimation unit and the output torque estimation unit of FIG. 15, and FIG. 17 is a diagram illustrating the power pulsation extraction unit of FIG. 18 is a diagram showing the inverter output voltage control unit of FIG. 15, and FIG. 19 is a diagram showing the torque control unit of FIG. In FIG. 15, the configuration is the same as that of the first embodiment except for the configuration of the operation mode determination unit 7 and the inverter output voltage control unit 12. The inverter control device 17 according to the second embodiment includes a coordinate converter 6, an operation mode determination unit 7, a rotation speed command value generation unit 13, an inverter output voltage control unit 12, and an inverter gate signal generation unit 14. The operation mode determination unit 7 according to the second embodiment includes a power estimation unit 8, a torque compensation value generation unit 25, and a determination signal generation unit 10. That is, the operation mode determination unit 7 of the second embodiment includes a torque compensation value generation unit 25 instead of the power pulsation extraction unit 9 of the first embodiment. The torque compensation value generation unit 25 includes an output torque estimation unit 35 and a torque pulsation extraction unit 36. As shown in FIG. 18, the inverter output voltage control unit 12 according to the second embodiment includes a current control unit 22, a voltage coordinate converter 23, a speed control unit 46, a torque command value compensation unit 47, a torque control unit 48, and magnetic flux estimation. A unit 18, an integrator 19, and a subtracter 20 are provided. In FIG. 15, only the current control unit 22, the voltage coordinate converter 23, the speed control unit 46, the torque command value compensation unit 47, and the torque control unit 48 are shown as the configuration of the inverter output voltage control unit 12.

The output torque estimation unit 35 calculates the torque (estimated torque τ ^) of the electric motor 1 based on the estimated power P ^ calculated by the power estimation unit 8. The torque pulsation extraction unit 36 extracts a pulsation component of power (torque pulsation component τ ^ kf) obtained by detecting the estimated torque τ ^ estimated by the output torque estimation unit 35 at a specific frequency. Further, the torque pulsation extraction unit 36 extracts the amplitude || τkf || of the pulsation component (ripple component) of torque obtained by detecting the estimated torque τ ^ at a predetermined frequency. The speed control unit 46 of the inverter output voltage control unit 12 generates a torque command τ * based on the rotation speed command ωe *. The torque command value compensation unit 47 receives a torque correction command τref, which is a compensation value corrected based on the torque command τ * generated by the speed control unit 46 and the torque pulsation component τ ^ kf extracted by the torque pulsation extraction unit 36. Generate. The torque control unit 48 generates a dq axis current command vector Idq * based on the corrected torque correction command τref. The voltage coordinate converter 23 calculates a voltage command vector Vuvw * for controlling the electric motor 1. The determination signal generation unit 10 of the operation mode determination unit 7 determines a compression operation mode of the compressor 80 based on the dominant pulsation component of the torque pulsation component τ ^ kf extracted by the torque pulsation extraction unit 36. Generate hnt. In the above configuration, the inverter control device 17 according to the second embodiment can cause the output torque of the electric motor 1 to follow the shaft torque, and can reduce fluctuations in the rotational speed of the electric motor 1.

As shown in FIG. 16, the output torque estimation unit 35 includes a divider 38 and a multiplier 37, and calculates the estimated torque τ ^ from the estimated power P ^ calculated by the power estimation unit 8. Based on the estimated power P ^ calculated by the power estimation unit 8, the number of pole pairs Pm of the electric motor 1, and the estimated rotational speed ω ^ e of the electric motor 1, the output torque estimating unit 35 calculates the estimated torque τ ^ by the equation (14). ).

The calculation formula for obtaining the estimated torque τ ^ from the estimated power P ^ is not limited to the expression (14), but the current of the motor 1 (d-axis current id, q-axis current iq as shown in the following expression (15)). ). However, (PHI) f is an armature linkage magnetic flux by a permanent magnet.

FIG. 17 shows the torque pulsation extraction unit 36 that calculates the torque pulsation component caused by the pressure fluctuation or the shaft torque fluctuation of the compressor 80 and the magnitude (amplitude) of the torque pulsation component. Similar to the power pulsation extraction unit 9 of the first embodiment, the torque pulsation extraction unit 36 extracts the amplitude || τkf || of the torque pulsation component (ripple component) obtained by detecting the estimated torque τ ^ at a predetermined frequency. . k is an integer from 1 to N, and the amplitude of the pulsating component exists from || τ1f || to || τNf ||. However, unlike the power pulsation extraction unit 9 of the first embodiment, the torque pulsation extraction unit 36 is a torque pulsation component τ ^ that is a temporal vibration component at a predetermined frequency (eg, fkf, k is an integer from 1 to N). kf is extracted as in the following equation (16). Torque pulsation components extracted by the torque pulsation extraction unit 36 exist from τ ^ 1f to τ ^ Nf.

As in the first embodiment, the torque pulsation extraction unit 36 extracts torque pulsation components corresponding to the number of compression units in order to determine the operation mode of the compressor 80, that is, the number of compression units being operated. As shown in FIG. 17, the torque pulsation extraction unit 36 will be described using an example in which the compressor 80 includes N compression units. The torque pulsation extraction unit 36 includes N torque pulsation component extractors. Each torque pulsation component extractor generates a torque pulsation component having a specific frequency and calculates the magnitude (amplitude) of the torque pulsation component. In FIG. 17, three torque

pulsation component extractors

101a, 101b, and 101c are shown. The code of the power pulsation component extractor uses 101 as a whole, and 101a, 101b, and 101c are used for distinction. The specific configuration of the torque pulsation component extractor 101 is shown in the torque pulsation component extractor 101b. The torque pulsation component extractor 101a calculates the torque pulsation component τ ^ 1f at the frequency f1f and its amplitude || τ1f ||. The torque pulsation component extractor 101b calculates the torque pulsation component τ ^ kf and its amplitude || τkf || at the frequency fkf. The torque pulsation component extractor 101c calculates the torque pulsation component τ ^ Nf and its amplitude || τNf || at the frequency fNf. The configuration of the torque pulsation component extractor 101 will be described using the torque pulsation component extractor 101b as an example.

The torque pulsation component extractor 101 includes a cosine wave generator 39, a sine wave generator 40,

integrators

42a and 42b,

multipliers

41a, 41b, 43a and 43b, an adder 44, and an amplitude calculator 102. A torque cosine wave component τkA, which is a cosine wave component of the estimated torque τ ^, is generated as follows. First, the cosine wave generation unit 39 and the multiplier 41a generate a cosine wave component of the estimated torque τ ^ of the frequency fkf, that is, an initial cosine wave component including noise. The integrator 42a integrates this initial cosine wave component, and calculates the amplitude || τkA || of the cosine wave component. The multiplier 43a multiplies the amplitude || τkA || by the cosine wave having the frequency fkf to generate a torque cosine wave component τkA from which noise has been removed. The torque cosine wave component τkA is the first term (cos term) on the right side of Equation (16).

The torque sine wave component τkB, which is the sine wave component of the estimated torque τ ^, is generated as follows. First, the sine wave generation unit 40 and the multiplier 41b generate a sine wave component of the estimated torque τ ^ of the frequency fkf, that is, an initial sine wave component including noise. The integrator 42b integrates this initial sine wave component to calculate the amplitude || τkB || of the sine wave component. The multiplier 43b multiplies the sine wave of amplitude || τkB || and frequency fkf to generate a torque sine wave component τkB from which noise has been removed. The torque sine wave component τkB is the second term (sin term) on the right side of Equation (16). Based on the adder 44, the torque cosine wave component τkA and the torque sine wave component τkB are integrated to generate a torque pulsation component τ ^ kf that is a temporal vibration component of the estimated torque τ ^ at a predetermined frequency fkf. The The torque pulsation component τ ^ kf is a combined torque pulsation component.

The amplitude calculation unit 102 calculates the amplitude || of the torque pulsation component τ ^ kf at the frequency fkf from the amplitude || τkA || of the cosine wave component and the amplitude || τkB || τkf || is calculated.

The principle of torque compensation will be described with reference to FIGS. FIG. 20 is a diagram for explaining a change in rotational speed due to torque pulsation, and FIG. 21 is a diagram for explaining compensation for torque pulsation according to Embodiment 2 of the present invention. First, the mechanical rotation speed ωm output from the electric motor 1 can be calculated as shown in Expression (18).

Here, p is a differential number, Te and TL are the output torque and load torque of the electric motor 1, respectively. J is the moment of inertia.

FIG. 20 shows a case where the output of the speed control unit 46 is not compensated, and FIG. 21 shows a case where the output of the speed control unit 46 is compensated. The upper part of FIG. 20 shows a torque command waveform 92 of the torque command τ *. The middle part of FIG. 20 shows an output torque waveform 94a of the output torque Te and a load torque waveform 93a of the load torque TL. The lower part of FIG. 20 shows a rotational speed waveform 95a of the mechanical rotational speed ωm of the electric motor 1. In FIG. 20, the horizontal axis represents time, and the vertical axis represents torque and rotational speed. 21 shows a torque command waveform 92 of the torque command τ * and a torque vibration component waveform 96 of the torque vibration component. 21 shows an output torque waveform 94b of the output torque Te and a load torque waveform 93b of the load torque TL. The lower part of FIG. 21 shows a rotational speed waveform 95b of the mechanical rotational speed ωm of the electric motor 1. In FIG. 21, the horizontal axis represents time, and the vertical axis represents torque and rotational speed.

In the speed controller 46, the estimated rotational speed ω ^ e of the electric motor 1 based on the magnetic flux (d-axis rotor magnetic flux Φ ^ dr) calculated by the magnetic flux estimator 18 is the rotational speed set by the rotational speed command value generator 13. The torque command τ * is calculated so as to be balanced with (the value of the rotation speed command ωe *). Therefore, if the rotational speed command ωe * is constant, the torque command τ * is constant. However, as shown in the middle part of FIG. 20, the difference Te−TL (the load torque waveform 93a and the output torque) between the output torque Te of the electric motor 1 controlled with a constant torque command value and the load torque TL in the operation mode of the compressor 80. The shaded area surrounded by the waveform 94a) changes periodically. For this reason, as shown in the lower rotation speed waveform 95a of FIG. 20, the pulsation of the rotation speed ωe increases. Therefore, it is necessary to reduce the difference Te−TL so that the output torque Te can follow the load torque TL, thereby reducing the pulsation of the rotational speed ωe. Therefore, as shown in FIG. 21, the output torque Te is balanced with the load torque TL based on the torque pulsation component τ ^ kf extracted from the load torque TL being superimposed on the torque command τ * as a compensation amount. The pulsation of the speed ωe can be reduced. The rotational speed amplitude of the rotational speed waveform 95a when the compensation is not performed is A1, the rotational speed amplitude of the rotational speed waveform 95b when the compensation is performed is A2, and the rotational speed amplitude A2 is greater than the rotational speed amplitude A1. Is also getting smaller.

The torque pulsation component τ ^ kf is a torque pulsation component having a frequency fkf, and the frequency is any one of f1f to fNf. The optimum frequency may be selected as follows. For example, the frequency of the dominant pulsation component, that is, the frequency at which the amplitude || τkf || In this case, the frequency fkf is the selected frequency.

The discrimination signal generator 10 has the same configuration as that shown in FIG. However, the maximum amplitude detection unit 33 of the determination signal generation unit 10 of the second embodiment uses the torque pulsation amplitudes (|| τ1f || ˜ ||) of the plurality of frequencies (f1f to fNf) calculated by the torque pulsation extraction unit 36. τNf || is input, the frequency having the maximum amplitude is detected, and the function value U calculated by changing the input of Equation (9) to || τ1f || Based on the function value U, the operation mode in which the compressor 80 operates is determined based on the determination unit 34 that determines based on the following language expression determination rule, and generates a determination signal hnt.

When U = mf, hnt = m. In this case, it determines with the operation mode which circulates a refrigerant | coolant with m compression parts. Here, m is an integer from 1 to N. Three cases are specifically shown below.
When U = 1f, hnt = 1. In this case, it determines with the operation mode which circulates a refrigerant | coolant by one compression part.
When U = kf, hnt = k. In this case, it is determined as an operation mode in which the refrigerant is circulated by the k compression units.
When U = Nf, hnt = N. In this case, it determines with the operation mode which circulates a refrigerant | coolant in all the compression parts.

Rotational speed command value generation unit 13 is provided with an operating speed command ωe * for each operating mode, as in the first embodiment. The rotation speed command value generation unit 13 outputs a predetermined rotation speed command ωe * in each operation mode based on the determination signal hnt output from the determination signal generation unit 10.

FIG. 18 shows a configuration of the inverter output voltage control unit 12 to which the principle of torque compensation shown in FIG. 21 is applied. In FIG. 18, the inverter output voltage control unit 12 of the second embodiment includes a speed control unit 46, a torque command value compensation unit 47, and a torque control unit 48 instead of the speed control unit 21 of the first embodiment. . The speed controller 46 receives the value of the rotational speed command ωe * generated by the rotational speed command value generator 13 and the speed difference Δω between the estimated rotational speed ω ^ e estimated by the magnetic flux estimator 18 and inputs the estimated rotational speed ω ^. A torque command τ * is generated and output so that e matches the value of the rotational speed command ωe *. Here, the speed difference Δω is ωe * −ω ^ e.

The torque command value compensation unit 47 adds the torque pulsation component τ ^ kf, which is the vibration component of the estimated torque τ ^ of the electric motor 1 calculated by the torque pulsation extraction unit 36, and the torque command τ * output by the speed control unit 46. The corrected torque correction command τref is output. The torque correction command τref can be expressed as Equation (19).

A general compensation method is to convert previously recorded load torque data into a torque command for the rotation angle of the electric motor 1. If this general compensation method is applied, the load torque fluctuates based on the rotation of the electric motor 1 of the compressor 80, so a lot of data on the load torque with respect to the rotation speed of the electric motor 1 must be stored. Further, in a general compensation method, since the load torque is different even at a constant rotational speed, it is difficult to accurately calculate the compensation amount. In the present invention, the pulsation of the load torque TL is estimated in real time, and the torque pulsation component τ ^ kf that becomes the compensation amount even when the load torque TL fluctuates can be calculated in real time. Therefore, data on the load torque TL is not necessary, and torque compensation can be performed accurately.

The torque control unit 48 calculates the dq-axis current command vector Idq * of the electric motor 1 so that the corrected torque correction command τref is input and the electric motor 1 outputs a torque having a value specified by the torque correction command τref. For example, when the existing maximum torque control is used, the d-axis current command id * and the q-axis current command iq * of the dq-axis current command vector Idq * can be calculated as shown in FIG. Then, the d-axis current command id * and the q-axis current command iq * are obtained from Expression (20). As shown in FIG. 19, the torque control unit 48 includes a current command value generation unit 49a, an angle command value generation unit 49b, a cosine calculation unit 50, a sine calculation unit 51, and

multipliers

52a and 52b.

The current command value generation unit 49a has a current torque map describing the relationship between the torque τ and the current vector Ia, and generates a current command vector Ia * from the torque correction command value τref. The angle command value generation unit 49b has an angle current map in which the relationship between the current vector Ia and the advance angle β from the q axis is described. The current command vector Ia * is converted into a command value β * of the advance angle from the q axis. Ask for. A d-axis current command id * is generated based on the sine calculation unit 51 and the multiplier 52a, and a q-axis current command iq * is generated based on the cosine calculation unit 50 and the multiplier 52b. However, as shown in FIG. 22, the angle β is an advance angle from the q-axis of the current vector Ia constituting the d-axis current id and the q-axis current iq. FIG. 22 is a diagram for explaining a current vector according to the second embodiment of the present invention.

It should be noted that the current torque map of the current command value generation unit 49a and the angle current map of the angle command value generation unit 49b may be obtained by simple analysis in advance based on the constants of the electric motor 1. Moreover, you may use the relationship calculated | required from the actual measurement value as an electric current torque map and an angle electric current map.

FIG. 23 is a diagram showing a vibration suppression result according to the second embodiment of the present invention. FIG. 23 shows the effect when the compressor 80 is controlled by the inverter output voltage control unit 12 and the inverter 16 using torque compensation, that is, the effect of reducing the vibration generated in the compressor 80. The horizontal axis of FIG. 23 is the frequency of pulsation of the rotational speed of the electric motor 1, and the vertical axis is the vibration level, that is, the amplitude (mm / s ^ 2) of the rotational speed pulsation vibration component. The vibration level of the compressor 80 was measured by attaching a uniaxial acceleration sensor to the main body of the compressor 80. In FIG. 23, the electric motor 1 is driven at a rotational speed command value of 2100 rpm (35 rps) when the compressor 80 is operated alone. A rotational speed waveform having a pulsating component of the frequency f1f and a frequency f2f twice that in the single operation is subjected to Fourier transform (FFT) analysis, and the vibration level with compensation is compared with the vibration level without compensation.

The

pulsation component characteristics

66 and 67 having a frequency of 35 Hz are the characteristics of the frequency f1f. The pulsation component characteristic 66 is a measurement result without suppression (no torque compensation), and the pulsation component characteristic 67 is a measurement result with suppression (torque compensation). Pulsating

component characteristics

68 and 69 having a frequency of 70 Hz are characteristics of the frequency f2f. The pulsation component characteristic 68 is a measurement result without suppression (no torque compensation), and the pulsation component characteristic 69 is a measurement result with suppression (torque compensation). The pulsation component characteristic 66 of the frequency f1f can be set to about ¼ of the pulsation component characteristic 67 by performing torque compensation. The pulsation component characteristic 68 of the frequency f2f can be set to about ¼ of the pulsation component characteristic 69 by performing torque compensation. Therefore, the inverter output voltage control unit 12 according to the second embodiment performs the torque compensation so that the pulsation having the main frequency f1f in the pulsation component of the estimated rotational speed ω ^ e of the electric motor 1 as shown in FIG. The component can be reduced to at least half or less based on no compensation.

Furthermore, the inverter output voltage control unit 12 according to the second embodiment adds the dominant shaft torque fluctuation (torque pulsation component τ ^ kf) caused by the machine of the compressor 80 as a compensation amount to the torque command τ *. The torque correction command τref, which is generated based on the control torque, can follow the load torque TL, and the rotational speed fluctuation (vibration) of the electric motor 1 can be reduced. Furthermore, the inverter output voltage control unit 12 according to the second embodiment calculates the torque compensation amount (torque pulsation component τ ^ kf) in real time, thereby increasing or decreasing the load fluctuation of the compressor 80 based on the rotational speed ωe of the electric motor 1. Even so, it is possible to follow the load torque TL without preparing a torque pattern. Furthermore, the inverter output voltage control unit 12 according to the second embodiment can reduce the pulsation of the output power and improve the efficiency of the compressor when performing control so as to obtain the power required by the load. is there.

In addition, since most of the conventional torque disturbances are calculated based on the fluctuation of the q-axis current, if the pressure of the compressor 80 or the shaft torque of the motor 1 fluctuates significantly, the q-axis current and the d-axis current Since both fluctuate, it cannot be calculated correctly. On the other hand, the inverter output voltage control unit 12 according to the second embodiment uses the pulsating component τ ^ kf having the largest amplitude from the estimated torque τ ^ obtained from the estimated power P ^ as a compensation amount to the torque command τ *. Since the torque correction command τref, which is a control torque, is generated by addition, the motor 1 of the compressor 80 is controlled based on the torque correction command τref, so that the control is performed regardless of the magnitude and speed of the load torque TL. Torque can be followed. Further, the inverter output voltage control unit 12 according to the second embodiment feeds back and corrects the deviation between the dq axis of the control command and the dq axis of the electric motor 1, so that an accurate torque can be calculated.

In the second embodiment, the rotation speed command value generation unit 13 is provided with the operation speed command ωe * for each operation mode as in the first embodiment. There is no need to set each operation mode. For example, the estimated power P ^ obtained by the power estimator 8 and the estimated torque τ ^ obtained by the output torque estimator 35, or the largest amplitude obtained by the maximum amplitude detector 33 of the discrimination signal generator 10 of the first embodiment. The driving speed command ωe * may be generated each time using the power fluctuation component P ^ kf having a torque pulsating component τ ^ kf having the largest amplitude obtained by the discrimination signal generation unit of the second embodiment. . Further, the refrigerant pressure difference condition inside the compressor, which is determined by the upper air conditioner control using the difference between the room temperature and the user set temperature, and the estimated power P ^, the estimated torque τ ^ or the power fluctuation component P ^ A driving power target value may be set from kf and the torque pulsation component τ ^ kf, and the driving speed command, that is, the rotational speed command ωe * may be determined so as to be the power.

Since the inverter compressor 100 according to the second embodiment includes the inverter control device 17 according to the second embodiment, the operation mode of the compressor 80 can be determined. Further, the inverter compressor 100 of the second embodiment automatically detects the switching of the operation mode in the compressor 80 in which the number of compression units for compressing the refrigerant changes, and the electric motor is matched with the mode-specific shaft torque variation pattern. By performing control so that the torque of 1 can be output, vibration suppression control that is not affected by the operation mode can be performed. However, the present invention can also be applied to a compressor having only one operation mode (for example, the single compressor or the scroll compressor of the first embodiment).

Embodiment 3 FIG.
FIG. 24 shows a compressor according to Embodiment 3 of the present invention. 25 is a schematic cross-sectional view of the first compression portion of FIG. 24, and FIG. 26 is a schematic cross-sectional view of the second compression portion of FIG. 27 is a block diagram showing an inverter control device and an inverter compressor according to Embodiment 3 of the present invention, and FIG. 28 is a diagram showing an operation mode switching detection unit and a torque pulsation extraction unit of FIG. In the third embodiment of the present invention, the compressor 80 is a twin rotary compressor. FIG. 24 shows a compressor 80 that is a two-cylinder compressor as a twin rotary compressor. The compressor 80 includes the electric motor 1, a first compression unit 83 a, a second compression unit 83 b, and a shaft 84.

The drive modes for driving the electric motor 1 of the compressor 80 are a single operation mode and a parallel operation mode. The single operation mode is an operation mode in which one of the two compression units (the first compression unit 83a and the second compression unit 83b) does not compress the refrigerant even when the shaft 84 of the electric motor 1 rotates. This is an operation mode in which the compression section of the compressor is compressed. The parallel operation mode includes an operation mode in which two compression units (first compression unit 83a and second compression unit 83b) are simultaneously compressed by shifting the refrigerant compression timing by 180 degrees. Accordingly, in the single operation mode, since only one compression unit is moving, the frequency of the power pulsation component is the same as the mechanical rotation frequency of the electric motor 1, that is, the frequency f1f. On the other hand, in the parallel operation, since the two compression parts are moving, the frequency of the power pulsation component is twice the mechanical rotation frequency of the electric motor 1, that is, the frequency f2f (2 × f1f). The twin rotary compressor automatically switches between single operation and parallel operation according to the internal differential pressure conditions.

The first compression portion 83a includes a piston 74a that moves as the shaft 84 rotates, a vane 76a, a spring 77a, a suction port 75a that sucks refrigerant, a discharge port 78a that discharges refrigerant, and an on-off valve 79a that opens and closes the discharge port 78a. Prepare. An arrow 88 a in FIG. 25 indicates the rotation direction of the shaft 84. An arrow 89 in FIG. 25 indicates the flow of gas discharged from the discharge port 78a. The second compression section 83b includes a piston 74b that moves with the rotation of the shaft 84, a vane 76b, a magnet 87 that applies an attractive magnetic force in a direction away from the piston 74b, a suction port 75b that sucks refrigerant, and a refrigerant. And an opening / closing valve 79b for opening and closing the discharge port 78b. The position of the piston 74a of the first compression section 83a (contact position between the piston 74a and the shaft 84) is 180 degrees with respect to the position of the piston 74b of the second compression section 83b (contact position between the piston 74b and the shaft 84). It's off. An arrow 88b in FIG. 26 indicates the rotation direction of the shaft 84.

25 is in a state during compression operation, and the second compression unit 83b in FIG. 26 is in a state where compression is stopped. In the first compression portion 83a during the compression operation of FIG. 25, the tip end side of the vane 76a is in contact with the piston 74a, and the rotation of the piston 74a is based on the inner wall of the first compression portion 83a, the piston 74a, and the vane 76a. A compression chamber 82 is formed. The refrigerant sucked from the suction port 75a is compressed in the compression chamber 82 and then discharged from the discharge port 78a. 26, the rear end side of the vane 76b is attracted and fixed to the magnet 87, the tip of the vane 76b is separated from the piston 74b, the inner wall of the second compression portion 83b, the piston 74b. The compression chamber is not formed based on the vane 76b, and the compression operation is not performed even if the piston 74b is rotated.

FIG. 27 shows the configuration of the inverter control device 17 that performs optimal control in the third embodiment. The inverter control device 17 according to the third embodiment controls the electric motor 1 of the compressor 80, which is a two-cylinder compressor whose operation mode can be automatically switched, via the inverter 16. The inverter control device 17 of the third embodiment is the same as that of the second embodiment except for the configuration of the operation mode determination unit 7. The operation mode discriminating unit 7 of the third embodiment replaces the torque compensation value generating unit 25 and the discrimination signal generating unit 10 in the operation mode discriminating unit 7 of the second embodiment, and an operation mode switching detecting unit 53, a torque pulsation extracting unit. 70. As illustrated in FIG. 28, the operation mode switching detection unit 53 includes an operation mode determination unit 60 that generates a determination signal hnt, similar to the determination signal generation unit 10 of the second embodiment.

The operation mode switching detection unit 53 determines whether the operation mode is the single operation or the parallel operation when the compression operation mode operated by the compressor 80 is switched based on the magnitude of the power pulsation component as shown in the first embodiment. To do. The torque pulsation extraction unit 70 extracts the torque pulsation component of the operation mode determined by the operation mode switching detection unit 53. With the above configuration, the inverter control device 17 according to the third embodiment is based on adding the extracted torque pulsation (torque pulsation component τ ^ kf) as a compensation amount to the torque command τ * output from the inverter output voltage control unit 12. The output torque of the electric motor 1 can follow the shaft torque of the compressor 80, which is a two-cylinder compressor, and fluctuations in the rotational speed ωe of the electric motor 1 can be reduced.

28, the operation mode switching detection unit 53 that determines the compression operation mode in which the compressor 80 operates and the torque pulsation extraction unit 70 that extracts the torque pulsation in the determined operation mode will be described in detail. The operation mode switching detection unit 53 includes a power pulsation extraction unit 61 and an operation mode determination unit 60 that determines the operation mode based on the pulsation component amplitude calculated by the power pulsation extraction unit 61. The power pulsation extraction unit 61 extracts a power pulsation component x1f and an amplitude || x1f || of the same frequency f1f as the motor machine frequency from the estimated power P ^ calculated by the power estimation unit 8; A discrete Fourier transform unit 63 that extracts the power pulsation component x2f and the amplitude || x2f || of the frequency f2f that is twice the frequency is provided. The frequency f1f is the motor machine frequency in the single operation, and the frequency f2f is the frequency of the pulsation component generated in the parallel operation. Discrete Fourier transform unit 62 detects pulsation components as shown in equations (5) to (8), and extracts power pulsation component x1f and its amplitude || x1f ||. Discrete Fourier transform unit 63 detects pulsation components as shown in equations (5) to (8), and extracts power pulsation component x2f and its amplitude || x2f ||.

The operation mode determination unit 60 determines the operation mode of the compressor 80 based on the dominant pulsation component in the pulsation component, and outputs a determination signal hnt. The operation mode determination unit 60 includes a maximum amplitude detection unit 33 and a determination unit 34. A method for determining whether the operation mode of the compressor 80 is the single operation or the parallel operation in the operation mode determination unit 60 will be described later.

The torque pulsation extraction unit 70 includes a switching unit 57 that switches between power pulsation extraction components and a torque pulsation generation unit 65 that extracts torque pulsation in the operation mode. The switching unit 57 switches the power pulsation component according to the operation mode of the compressor 80 based on the determination signal hnt. The switching unit 57 outputs the power pulsation component x1f to the torque pulsation generation unit 65 when the operation mode is a single operation, and outputs the power pulsation component x2f to the torque pulsation generation unit 65 when the operation mode is parallel operation. The power pulsation component x1f is connected to the terminal 54 of the switching unit 57, and the power pulsation component x2f is connected to the terminal 55. The selected power pulsation component is output from the output terminal 56 of the switching unit 57 to the torque pulsation generation unit 65.

The torque pulsation generator 65 includes a divider 58 and a multiplier 59. The divider 58 divides the input power pulsation component by the estimated rotational speed ω ^ e calculated by the magnetic flux estimator 18 (see FIG. 18). The multiplier 59 multiplies the input by Pm times that is the number of pole pairs of the electric motor 1. The torque pulsation generator 65 generates a torque pulsation component τ ^ kf corresponding to the operation mode. Here, k is 1 or 2. Therefore, the torque pulsation generator 65 generates a torque pulsation component τ ^ 1f when the operation mode is a single operation, and generates a torque pulsation component τ ^ 2f when the operation mode is a parallel operation. In FIG. 27, the estimated rotational speed ω ^ e output from the magnetic flux estimator 18 as the power pulsation component of the inverter output voltage controller 12 is omitted.

A determination method of the operation mode determination unit 60 will be described. Similar to the discrimination signal generation unit 10 of the first embodiment, the operation mode discrimination unit 60 performs compression based on the amplitude || x1f || of the power pulsation component x1f and the amplitude || x2f || of the power pulsation component x2f. The operation mode in which the machine 80 is currently operating is determined. The determination signal hnt output from the operation mode determination unit 60 has various patterns, but may be expressed as follows.
When the operation mode is single operation, hnt = 1
When the operation mode is parallel operation, hnt = 2

The maximum amplitude detection unit 33 of the operation mode determination unit 60 is the same as the maximum amplitude detection unit 33 of the determination signal generation unit 10 of the first embodiment, and the two input amplitudes || x1f || and || x2f | The frequency having the maximum amplitude is detected from | and output as a function value U. Based on the function value U, the operation mode in which the compressor 80 operates is determined based on the determination unit 34 that determines based on the following language expression determination rule, and generates a determination signal hnt.
When U = 1f, hnt = 1. In this case, it is determined that the operation mode is a single operation mode in which the refrigerant is circulated by one compression unit.
When U = 2f, hnt = 2 is set. In this case, it is determined that the operation mode is a parallel operation in which the refrigerant is circulated by the two compression units.

The terminal connection conditions in the switching unit 57 are shown below.
When hnt = 1, the output terminal 56 is connected to the terminal 54. When hnt = 2, the output terminal 56 is connected to the terminal 55.

The determination signal hnt generated by the operation mode switching detection unit 53 is output to the torque pulsation extraction unit 70 and the rotation speed command value generation unit 13. In the operation mode switching detection unit 53, the switching unit 57 switches to the torque pulsation component in the compression operation mode according to the determination signal hnt. Then, the rotational speed command value generation unit 13 outputs a predetermined rotational speed command ωe * in each operation mode based on the determination signal hnt.

Further, the torque command value compensation unit 47 corrects based on the addition of the torque pulsation component τ ^ kf based on the discrimination signal hnt as the compensation amount based on the discrimination signal hnt to the torque command τ * generated by the speed control unit 46. The torque correction command τref is output. Based on the correction command τref, the torque control unit 48 calculates a dq-axis current command vector Idq * of the electric motor 1 (see FIG. 18). Thereafter, the electric motor 1 of the compressor 80 is controlled via the current controller 22, the voltage coordinate converter 23, the inverter gate signal generator 14, and the inverter 16. The inverter control device 17 according to the third embodiment can control the electric motor 1 so that the output torque Te is balanced with the load torque TL based on the correction command τref, and can reduce the pulsation generated in the rotational speed ωe of the electric motor 1.

Therefore, the inverter control device 17 according to the third embodiment automatically determines the operation mode of the twin rotary compressor (compressor 80) so that a plurality of operation modes can be automatically switched, and loads the output torque Te based on the determination. Based on following the torque TL, pulsation of the rotational speed ωe of the electric motor 1 can be reduced, and stable operation of the electric motor 1 can be realized. Further, the inverter control device 17 according to the third embodiment can reduce the pulsation of the rotational speed ωe of the electric motor 1 and realize a stable operation of the electric motor 1. Therefore, the electric motor 1 can be operated with low noise and low vibration, and the compressor Optimal control at 80 can be achieved.

Further, the inverter control device 17 according to the third embodiment immediately determines the compression operation mode and switches the pressure fluctuation when the compressor 80 having the function of automatically switching the compression operation mode is switched to the compression operation mode having a different pressure fluctuation. Is reflected in the electric power of the electric motor 1, so that optimal control can be performed according to the operation mode.

As described above, if one of the two compression sections of the twin rotary compressor (compressor 80) performs a single operation in which the refrigerant is compressed, the refrigerant circulation rate is halved. Therefore, although the compressor 80 is normally operated in parallel, in order to output the refrigerant circulation amount equivalent to that in the parallel operation in the single operation of the compressor 80, the rotational speed ωe of the electric motor 1 is set to 2 in the parallel operation. Need to double. The inverter control device 17 according to the third embodiment doubles the rotational speed ωe of the electric motor 1 when the compressor 80 is operated alone, so that the efficiency of the compressor 80 can be improved. . Further, the inverter control device 17 according to the third embodiment sets the rotational speed ωe to 1 at the time of the single operation so that the electric power becomes constant when the single operation is switched to the parallel operation in which the refrigerant is compressed by the two compression units. Must be doubled.

Examples of switching the rotational speed command ωe * between the single operation and the parallel operation as described above are shown in FIGS. FIGS. 29 and 30 are diagrams showing a control example of the compressor according to the third embodiment of the present invention. FIG. 29 shows a case where the compressor 80 is switched from parallel operation to single operation, and FIG. 30 is a case where the compressor 80 is switched from single operation to parallel operation. 29 and 30, the horizontal axis represents time, and the vertical axis represents the signal value or the command frequency of the rotational speed command ωe *. A waveform 71 is a parallel operation signal, and a waveform 72 is a single operation signal. A waveform 73 is a command frequency of the rotation speed command ωe *. In the present embodiment, since the operation mode is detected from the estimated power P ^ of the electric motor 1 by the operation mode switching detection unit 53, the parallel operation signal 71 and the individual operation signal 72 are not generated. It was described to distinguish between parallel and parallel operation.

As shown in FIG. 29, when the compressor 80 automatically switches from parallel operation to single operation at time t1, that is, when the parallel operation signal 71 changes from 1 to 0 and the single operation signal 72 changes from 0 to 1, The mode switching detection unit 53 determines the operation mode from the estimated power P ^ of the electric motor 1. When the operation mode switching detection unit 53 confirms the operation mode, it immediately outputs a determination signal hnt to the rotation speed command value generation unit 13. Upon receiving the determination signal hnt, the rotation speed command value generation unit 13 immediately outputs a command for doubling the rotation speed of the parallel operation to the speed control unit 46. That is, when the rotational speed command value generation unit 13 receives the determination signal hnt, the rotational speed command value ωe * is changed like the command frequency 73. In FIG. 29, it is described that the command frequency 73 changes from the time t1, but in reality, some delay time occurs.

As shown in FIG. 30, when the compressor 80 automatically switches from single operation to parallel operation at time t1, that is, when the parallel operation signal 71 changes from 0 to 1 and the single operation signal 72 changes from 1 to 0, the operation is started. The mode switching detection unit 53 determines the operation mode from the estimated power P ^ of the electric motor 1. When the operation mode switching detection unit 53 confirms the operation mode, it immediately outputs a determination signal hnt to the rotation speed command value generation unit 13. Upon receiving the determination signal hnt, the rotation speed command value generation unit 13 immediately outputs a command to halve the rotation speed during the single operation to the speed control unit 46. That is, when the rotational speed command value generation unit 13 receives the determination signal hnt, the rotational speed command value ωe * is changed like the command frequency 73. In FIG. 30, it is described that the command frequency 73 changes from the time t1, but in reality, some delay time occurs.

Therefore, the inverter control device 17 according to the third embodiment automatically determines the operation mode of the twin rotary compressor (compressor 80), and based on this, the rotational speed is set to 2 when switching from parallel operation to single operation. The rotation speed command ωe * to be doubled is set, and the rotation speed command ωe * to set the rotation speed to ½ is set in the case of switching from single operation to parallel operation. The amount of circulating refrigerant can be stabilized, the output power can be made constant, and the efficiency of the compressor 80 can be prevented from decreasing.

Since the inverter compressor 100 of the third embodiment includes the inverter control device 17, the operation mode of the compressor 80 can be determined. Further, the inverter compressor 100 according to the third embodiment automatically detects the switching of the operation mode in the compressor 80 in which the number of compression units for compressing the refrigerant is changed, and the electric motor is matched with the shaft torque variation pattern specific to the mode. By performing control so that the torque of 1 can be output, vibration suppression control that is not affected by the operation mode can be performed.

The inverter control device 17 according to the first to third embodiments has been described with reference to the example in which the compressor 80 is controlled. The present invention can also be applied to a machine having an operation mode having a component frequency.

Embodiment 4 FIG.
FIG. 31 is a block diagram showing an inverter control device and an inverter compressor according to the fourth embodiment of the present invention. FIG. 32 is a Bode diagram of the advanced phase filter in the phase adjustment unit of FIG. The Bode diagram of FIG. 32 shows the transfer characteristics of the filter of the phase adjustment unit 190. The inverter control device 17 according to the fourth embodiment differs from the inverter control device 17 according to the second embodiment in that the torque compensation value generation unit 25 includes a phase adjustment unit 190. In FIG. 31, the configuration other than the phase adjustment unit 190 is the same as that of the inverter control device 17 of the second embodiment.

31, the output torque estimation unit 35 of FIG. 31 calculates the torque of the electric motor 1 based on the estimated power P ^ calculated by the power estimation unit 8 as in the second embodiment. The torque pulsation extraction unit 36 extracts a pulsation component of electric power obtained by detecting the estimated torque τ ^ estimated by the output torque estimation unit 35 at a specific frequency. Further, the torque pulsation extracting unit 36 extracts the pulsation component amplitude of the torque detected from the estimated torque τ ^ or the phase-adjusted adjusted estimated torque τ ^ u at a predetermined frequency. Further, the torque command value compensation unit 47 includes a torque pulsation component τ ^ kf that is an amplitude component of the estimated torque τ ^ of the electric motor 1 calculated by the torque pulsation extraction unit 36, and a torque command τ * output by the speed control unit 46. By outputting a torque correction command τref corrected by adding, the output torque of the electric motor 1 can be made to follow the shaft torque. The inverter control device 17 according to the fourth embodiment can reduce the pulsation of the rotational speed of the electric motor 1 as shown in FIG. 23 according to the second embodiment.

However, when performing speed control, magnetic flux observer, or the like, control response or processing delay of the control device occurs, so the estimated torque τ ^ of the motor 1 calculated by the output torque estimating unit 35 is a corrected torque command correction. The phase of the estimated torque τ is delayed from the load torque. Therefore, by suppressing the phase delay in the estimated torque, specifically, by using the adjusted estimated torque τ ^ u adjusted so that the phase of the estimated torque τ ^ advances, the output torque can be brought close to the load torque. Thus, it is possible to improve the effect of reducing fluctuations in the rotational speed of the electric motor 1.

In order to suppress the phase delay of the estimated torque τ ^, the estimated torque τ ^ is calculated by the output torque estimator 35 and input to the phase adjuster 190 as shown in FIG. Specifically, the phase adjustment unit 190 employs a leading phase filter as indicated by the transfer function of Expression (21). The phase adjustment unit 190 outputs the adjustment estimated torque τ ^ u adjusted so that the phase advances as in Expression (22).

Here, s is a complex number. T1 and T2 are filter time constants, and are defined as in Expression (23).

ω1 and ω2 are angular frequencies [rad / s] specified by the filter, and are determined by the operating frequency range of the electric motor 1.

32 shows the transmission characteristics (amplitude and phase) of the adjusted estimated torque τ ^ u and the estimated torque τ ^ with respect to the torque correction command τref. The upper part of FIG. 32 shows amplitude transfer characteristics, and the lower part of FIG. 32 shows phase transfer characteristics. In the upper part of FIG. 32, the horizontal axis represents angular frequency [rad / s], and the vertical axis represents amplitude gain [dB]. In the lower part of FIG. 32, the horizontal axis represents angular frequency [rad / s], and the vertical axis represents phase [deg]. The phase characteristic 191 and the amplitude characteristic 193 are characteristics when there is a filter, that is, the characteristics of the adjustment estimated torque τ ^ u adjusted by the phase adjustment unit 190. The phase characteristic 192 and the amplitude characteristic 194 are characteristics when there is no filter, that is, characteristics of the estimated torque τ ^ not adjusted by the phase adjustment unit 190.

As shown in FIG. 32, in the case of no filter, the estimated torque τ ^ in the range of ω1 to ω2, which is the specified operating frequency range of the motor 1, for example, from the mechanical rotation angular frequency of 60 rad / s to 300 rad / s. It was found that the phase characteristic 192 is delayed. On the other hand, when there is a filter, the phase characteristic 191 of the adjustment estimated torque τ ^ u adjusted by the phase adjustment unit 190 has a suppressed delay. Therefore, by appropriately designing the angular frequency ω1 and the angular frequency ω2 in the leading phase filter of the phase adjusting unit 190, the delay of the estimated torque can be suppressed in the operating frequency range of the electric motor 1. That is, in the operating frequency range of the electric motor 1, the output torque is made close to the load torque by using the adjusted estimated torque τ ^ u adjusted so that the phase of the estimated torque τ ^ advances, thereby changing the rotational speed of the electric motor 1. It is possible to improve the reduction effect.

33 and 34 show the vibration suppression result according to the fourth embodiment of the present invention, that is, the effect of reducing the pulsation of the rotational speed of the electric motor 1. FIG. 33 is a diagram showing a vibration suppression result when there is no leading phase filter according to the fourth embodiment of the present invention, and FIG. 34 shows a vibration suppression result when there is a leading phase filter according to the fourth embodiment of the present invention. FIG. The horizontal axis of FIGS. 33 and 34 is the frequency order of the pulsation of the rotational speed of the electric motor 1, and the vertical axis of FIGS. 33 and 34 is the vibration level, that is, the amplitude [rpm] of the rotational speed pulsation vibration component. 33 and 34 show a Fourier transform (FFT) of the rotational speed waveform of the electric motor 1 having a pulsating component when the electric motor 1 is driven at a rotational speed command value of 1638 rpm (27.3 rps) when the compressor 80 is operated independently. It is the result of having compared the vibration suppression effect.

FIG. 33 is an FFT of the rotational speed of the electric motor 1 when the torque command is compensated and the vibration is suppressed by using the estimated torque value τ ^ without the leading phase filter. In FIG. 33,

pulsation component characteristics

195 and 196 are pulsation component characteristics when the frequency is 27.3 Hz, and are characteristics of the frequency f1f. The pulsation component characteristic 195 is a result without suppression, and the pulsation component characteristic 196 is a result with suppression. In FIG. 33, by performing torque correction on the pulsation component characteristic 195 of the frequency f1f, the amplitude of the pulsation component of the frequency f1f can be reduced from 155 rpm to 41 rpm. The

pulsation component characteristics

197 and 198 are pulsation component characteristics at a frequency twice as high as the frequency f1f, and are characteristics of the frequency f2f. The

pulsation component characteristics

199 and 200 are pulsation component characteristics at a frequency three times the frequency f1f, and are the characteristics of the frequency f3f.

FIG. 34 is an FFT of the rotational speed of the electric motor 1 when the torque command is compensated by using the estimated adjustment torque τ ^ u with a leading phase filter to suppress the vibration. In FIG. 34,

pulsation component characteristics

201 and 202 are pulsation component characteristics when the frequency is 27.3 Hz, and are characteristics of the frequency f1f. The pulsation component characteristic 201 is a result without suppression, and the pulsation component characteristic 202 is a result with suppression. In FIG. 34, by performing torque correction on the pulsation component characteristic 201 of the frequency f1f, the amplitude of the pulsation component of the frequency f1f can be reduced from 155 rpm to 20 rpm. The

pulsation component characteristics

203 and 204 are pulsation component characteristics at a frequency twice as high as the frequency f1f, and are characteristics of the frequency f2f. The

pulsation component characteristics

205 and 206 are pulsation component characteristics at a frequency three times the frequency f1f, and are the characteristics of the frequency f3f.

In the inverter control device 17 of the fourth embodiment, as shown in FIGS. 33 and 34, the adjusted estimated torque τ ^ u adjusted by using a leading phase filter, that is, adjusted so that the phase of the estimated torque τ ^ advances. By using, it is possible to avoid a correction deviation due to an estimated delay of the estimated torque τ ^, and to further reduce the amplitude of the rotational speed pulsation component of the electric motor 1 due to the load torque fluctuation in the compressor 80.

Embodiment 5 FIG.
FIG. 35 is a block diagram showing an inverter control device and an inverter compressor according to the fifth embodiment of the present invention. 36 is a diagram illustrating the torque pulsation extracting unit of FIG. 35, and FIG. 37 is a diagram illustrating the cosine wave generating unit and the sine wave generating unit of FIG. The inverter control device 17 according to the fifth embodiment is different from the inverter control device 17 according to the second embodiment in the configuration of the torque pulsation extraction unit 180 in the torque compensation value generation unit 25. In FIG. 35, the configuration other than that of the torque pulsation extraction unit 180 is the same as that of the inverter control device 17 of the second embodiment.

The purpose of the fifth embodiment is the same as that of the fourth embodiment, and is to avoid the delay of the estimated torque in the inverter control. However, in the inverter control device 17 of the fourth embodiment, there is a possibility that the phase of the estimated torque is adjusted and the amplitude (gain) of the estimated torque varies in the adopted advanced phase filter. Therefore, in the fifth embodiment, the torque pulsation extraction unit 180 includes the learning unit 111, and the phase amount θc of the torque pulsation component τ ^ kf is determined based on the phase amount (initial phase) θc output from the learning unit 111. Thus, it is possible to avoid the estimation delay of the torque pulsation component τ ^ kf.

The torque pulsation extraction unit 180 shown in FIG. 36 has a learning unit 111 added to the torque pulsation extraction unit 36 (see FIG. 17) of the second embodiment, and the phase amount θc is added to the cosine wave generation unit and the sine wave generation unit. Is configured to be input. 36, the torque pulsation component extractor 101b is one of the N torque pulsation component extractors 101 constituting the torque pulsation extraction unit 180, similarly to the torque pulsation component extractor 101b of FIG. The torque pulsation extraction unit 180 includes a learning unit 111, and the phase amount θc output from the learning unit 111 is mainly used by the cosine wave generation unit 159 and the sine wave generation unit 160 in the N torque pulsation component extractors 101. . FIG. 36 shows three torque

pulsation component extractors

101a, 101b, and 101c as in FIG. The specific configuration of the torque pulsation component extractor 101 is shown in the torque pulsation component extractor 101b.

The torque pulsation component extractor 101 includes a cosine wave generator 159, a sine wave generator 160,

integrators

42a and 42b,

multipliers

41a, 41b, 43a and 43b, an adder 44, and an amplitude calculator 102. A torque cosine wave component τkA, which is a cosine wave component of the estimated torque τ ^, is generated as follows. First, the cosine wave generation unit 159 and the multiplier 41a generate a cosine wave component of the estimated torque τ ^ of the frequency fkf, that is, an initial cosine wave component including noise. The integrator 42a integrates this initial cosine wave component, and calculates the amplitude || τkA || of the cosine wave component. The multiplier 43a multiplies the amplitude || τkA || by the cosine wave having the frequency fkf to generate a torque cosine wave component τkA from which noise has been removed. The torque cosine wave component τkA is the first term (cos term) on the right side of Equation (24).

The torque sine wave component τkB, which is the sine wave component of the estimated torque τ ^, is generated as follows. First, the sine wave generation unit 160 and the multiplier 41b generate a sine wave component of the estimated torque τ ^ of the frequency fkf, that is, an initial sine wave component including noise. The integrator 42b integrates this initial sine wave component to calculate the amplitude || τkB || of the sine wave component. The multiplier 43b multiplies the sine wave of amplitude || τkB || and frequency fkf to generate a torque sine wave component τkB from which noise has been removed. The torque sine wave component τkB is the second term (sin term) on the right side of Equation (24). Based on the adder 44, the torque cosine wave component τkA and the torque sine wave component τkB are integrated to generate a torque pulsation component τ ^ kf that is a temporal vibration component of the estimated torque τ ^ at a predetermined frequency fkf. The The torque pulsation component τ ^ kf is a combined torque pulsation component.

The expression (24) for generating the torque pulsation component τ ^ kf in the fifth embodiment is different from the expression (16) for generating the torque pulsation component τ ^ kf described in the second embodiment in terms of the cos function and the sin function. It is an equation that introduces the quantity θc. The torque pulsation component extractor 101 integrates the generated torque cosine wave component τkA and torque sine wave component τkB by the adder 44 based on the cosine wave component τkA and sine wave component τkB of the estimated torque τ ^. A pulsation component τ ^ kf of the estimated torque τ ^ at a predetermined frequency fkf is generated.

Specific configurations of the cosine wave generation unit 159 and the sine wave generation unit 160 are as follows. As illustrated in FIG. 37, the cosine wave generation unit 159 includes a multiplier 105, an adder 109, and a cosine wave function 107. The cosine wave generation unit 159 calculates an input angle to be input to the cosine wave function 107 based on the predetermined frequency fkf and the phase amount θc output from the learning unit 111, and passes the calculated input angle to the cosine wave function 107. Thus, a cos component at a predetermined frequency fkf is generated. Here, the input angle is 2 × π × fkf × t + θc, and the cos component is cos (2 × π × fkf × t + θc).

Similarly, as shown in FIG. 37, the sine wave generation unit 160 includes a multiplier 106, an adder 110, and a sine wave function 108. The sine wave generation unit 160 calculates an input angle to be input to the sine wave function 108 based on the predetermined frequency fkf and the phase amount θc output from the learning unit 111, and passes the calculated input angle to the sine wave function 108. Thus, a sin component at a predetermined frequency fkf is generated. Here, the input angle is 2 × π × fkf × t + θc, and the sin component is sin (2 × π × fkf × t + θc).

The operation of the learning unit 111 outputting the phase amount θc will be described with reference to FIG. FIG. 38 is a diagram showing a flow chat of the learning unit in FIG. First, the initial value of the phase amount θc is set to 0 (step S001). Next, in step S002, an estimated rotational speed ω ^ e that is an estimated value of the rotational speed of the electric motor 1 calculated by the magnetic flux estimating unit 18 of the inverter output voltage control unit 12 shown in FIG. 18 is read. A rotational speed difference ωerr, which is a difference between the estimated rotational speed ω ^ e of the electric motor 1 and the rotational speed command value ωe * output from the rotational speed command value generation unit 13, is calculated by Expression (25). 35, 36, and 37, the estimated rotational speed ω ^ e and the rotational speed command value ωe * input to the learning unit 111 of the torque pulsation extracting unit 180 are omitted.

Next, based on the calculated current rotational speed difference ωerr of the sampling and the stored rotational speed difference ωerrrp of the previous sampling, a fluctuation amount Δωerr of the rotational speed difference is calculated by Expression (26). Also, the absolute value | Δωerr | of the fluctuation amount Δωerr is calculated.

The fluctuation amount Δωerr of the rotational speed difference calculated by the equation (26) is affected by the magnitude of the pulsation of the rotational speed of the electric motor 1.

Next, in step S003, the absolute value | Δωerr | of the fluctuation amount of the rotational speed difference calculated in step S002 is compared with the designated threshold value ε1. If it is determined in step S003 that the absolute value | Δωerr | of the fluctuation amount of the rotational speed difference is ε1 or more (when the determination result is No), the process proceeds to step S004. In step S004, a new phase amount θc is generated by adding the positive increment δθ designated to the phase amount θc as shown in equation (27).

However, the phase amount θc is limited with the predetermined phase amount θcmax as an upper limit.

If it is determined in step S003 that the absolute value | Δωerr | of the fluctuation amount Δωerr of the rotational speed difference is smaller than ε1 (when the determination result is Yes), it is not necessary to adjust the phase amount θc, and the process proceeds to step S005. In step S005, learning unit 111 outputs phase amount θc to adder 109 of cosine wave generation unit 159 and adder 110 of sine wave generation unit 160. Torque pulsation component extractor 101 calculates torque pulsation component τ ^ kf based on equation (24) using the output phase amount θc. In step S006, after outputting the torque pulsation component τ ^ kf, if the sampling of the estimated rotational speed ω ^ e is completed, the learning operation is terminated. If the sampling is continued, the process returns to step S002, and the next sampling is performed. Repeat the same process as above.

FIG. 39 is a diagram showing a waveform of the command ω ^ e for the rotational speed of the electric motor. The horizontal axis is time, and the vertical axis is amplitude. A broken line 116 is a command value of the rotational speed command ωe *, and a waveform 115 is a waveform of the estimated rotational speed ω ^ e. As shown in FIG. 39, it can be seen that the estimated rotation speed ω ^ e of the electric motor 1 tends to decrease in amplitude fluctuation by the learning unit 111. In FIG. 39, the fluctuation amount Δωerr of four rotation speeds calculated for each sampling period Δts in the same phase in each cycle of the estimated rotational speed ω ^ e, that is, Δωerr1, Δωerr2, Δωerr3, and Δωerr4 are shown. If the fluctuation amounts Δωerr1, Δωerr2, Δωerr3, and Δωerr4 of the rotation speed difference for four times are observed, it can be seen that the fluctuation amount Δωerr of the rotation speed difference tends to gradually decrease.

The inverter control device 17 according to the fifth embodiment adjusts the phase of the torque pulsation component so that the learning unit 111 adjusts the phase of the torque pulsation component so that the fluctuation amount Δωerr of the rotational speed difference of the electric motor 1 falls below the specified threshold. The delay can be eliminated. Further, when the command value of the rotational speed command ω ^ e of the electric motor 1 is changed, the inverter 111 of the fifth embodiment causes the learning unit 111 to change the rotational speed command ω ^ e according to the command value of the rotational speed command ω ^ e. Appropriate phase amount can be adjusted in real time. Moreover, since learning unit 111 outputs phase amount θc adjusted in real time every time estimated rotation speed ω ^ e is read, inverter control device 17 of the fifth embodiment needs to set the phase amount in advance. In addition, it is possible to reduce the adjustment time in advance in the production of the inverter control device 17.

Embodiment 6 FIG.
FIG. 40 is a block diagram showing an inverter control device and an inverter compressor according to the sixth embodiment of the present invention. 41 is a diagram showing the torque command value switching unit of FIG. 40, and FIG. 42 is a diagram showing the torque learning unit of FIG. FIG. 43 is a diagram showing a configuration of the motor phase estimation unit of FIG. The inverter control device 17 according to the sixth embodiment changes the torque command value compensation unit 47 in the inverter control device 17 according to the second embodiment to the torque command value switching unit 120, and performs the interface switch 122, the torque learning unit 121, and the motor phase estimation. It differs from inverter control device 17 of Embodiment 2 in that unit 150 is provided. In FIG. 40, the configuration is the same as that of the inverter control device 17 of the second embodiment except for the configuration of the interface switch 122, the torque command value switching unit 120, the torque learning unit 121, and the motor phase estimation unit 150.

As shown in FIG. 43, the motor phase estimation unit 150 includes a multiplier 151 and an integrator 152, and the rotor phase of the motor 1 is determined based on the estimated rotational speed ω ^ e of the motor 1 calculated by the magnetic flux estimation unit 18. Estimate the rotation angle. The estimated machine rotation speed ω ^ m obtained by estimating the machine rotation speed ωm of the electric motor 1 is calculated by the equation (28).

However, Pm is the number of pole pairs of the electric motor 1.

The estimated mechanical rotational speed ω ^ m obtained by multiplying the estimated rotational speed ω ^ e by 1 / Pm by the multiplier 151 is input to the integrator 152. The estimated rotation angle θ ^ m of the rotor of the electric motor 1 is calculated as shown in Expression (29) based on the integrator 152. The estimated rotation angle θ ^ m is also referred to as an estimated machine rotation angle θ ^ m as appropriate.

The interface switch 122 determines the timing for suppressing the pulsation of the rotation speed of the electric motor 1. Specifically, the interface switch 122 outputs 1 when the vibration suppression is performed, and the interface switch 122 outputs 0 when the vibration suppression is stopped. The determination of whether to suppress or stop vibration may be performed automatically or manually. For example, in the following case, the operation for stopping the vibration suppression being performed and the operation for re-execution are automatically performed. Thereby, even when vibration suppression is difficult to work, the effect of vibration suppression can be increased by re-learning.

There are the following two conditions for determining whether to stop vibration suppression automatically.
(1) When the pulsation amplitude of the estimated rotational speed ω ^ e of the electric motor 1 exceeds a specified threshold value, a determination is made to stop vibration suppression.
(2) Based on the fluctuation of the determination signal hnt in the operation switching of the compressor 80, a determination is made to stop the vibration suppression.

On the other hand, when it is determined to manually stop vibration suppression, for example, vibration suppression is performed when the user selects the “static” mode on the remote controller.

In FIG. 41, the torque command value switching unit 120 switches the torque command value in the electric motor 1 based on the switching signal swt. Specifically, when the switching signal swt generated by the torque learning unit 121 is 0, the torque command value switching unit 120 connects the terminal 124 and the output terminal 126. In this case, the torque correction command τref is the torque command τ * calculated by the speed control unit 46. That is, it becomes like Formula (30).

On the other hand, when the switching signal swt is 1, the torque command value switching unit 120 connects the terminal 125 and the output terminal 126. In this case, the torque correction command τref is the torque command (corrected torque command) τ ** output from the torque learning unit 121. That is, the equation (31) is obtained.

That is, when the switching signal swt is 0, the torque command τ * calculated by the speed control unit 46 is input as a torque command value to the torque control unit 48 that performs feedback control. That is, when the switching signal swt is 0, the torque correction command τref becomes the torque command τ *, so that torque correction is not performed. On the other hand, when the switching signal swt is 1, the torque command value τ ** whose phase is corrected by the torque learning unit 121 is input to the torque control unit 48. That is, when the switching signal swt is 1, torque correction is performed based on the torque command value τ ** whose phase is corrected by the torque learning unit 121.

FIG. 42 shows a configuration of the torque learning unit 121. 42, the torque learning unit 121 includes a torque switch 128 and a learning algorithm processing unit 127 that control input of the estimated torque τ ^. The learning algorithm processing unit 127 includes an angle storage unit 142 and an estimated torque storage unit 143. The angle storage unit 142 stores the estimated mechanical rotation angle θ ^ m of the electric motor 1, and the estimated torque storage unit 143 stores the estimated torque τ ^ corresponding to the estimated mechanical rotation angle θ ^ m of the electric motor 1. 44 is a diagram illustrating the angle storage unit and the estimated torque storage unit of FIG. 42, and FIG. 45 is a diagram in which data is recorded in the angle storage unit and the estimated torque storage unit of FIG.

Whether the torque learning unit 121 stores the estimated torque τ ^ for the estimated mechanical rotation angle θ ^ m of the motor 1 calculated by the motor phase estimation unit 150 according to the state of the interface switch 122, or adjusts the phase of the estimated torque based on learning After that, it is determined whether to output the estimated torque τ ^ for θ ^ m.

The learning algorithm processing unit 127 is performed according to the flowchart of FIG. FIG. 46 is a diagram showing a flow chat of the learning algorithm processing unit of FIG. 47 is a diagram showing a flow chat of the recording mode execution process of FIG. 46, and FIG. 48 is a diagram showing a flow chat of the output mode execution process of FIG. First, the state of the interface switch (IF switch) 122 is monitored (step S101). If the interface switch is OFF, the process waits in step S101. If the interface switch is ON, the process proceeds to step S102, and recording mode execution processing is performed.

47. In step S201, the torque switch (TQ switch) 128 is turned on and the index i is set to 1 as shown in FIG. In step S202, θ ^ mi that is the estimated mechanical rotation angle θ ^ m of the electric motor 1 at the index i and τ ^ i that is the estimated torque τ ^ at the index i corresponding to the estimated mechanical rotation angle θ ^ mi. And stored in the angle storage unit 142 and the estimated torque storage unit 143 (estimated torque storage procedure).

The angle storage unit 142 and the estimated torque storage unit 143 respectively include N storage areas corresponding to a predetermined resolution N (for example, N = 360) from 0 ° to 360 ° of one rotation of the motor 1. The angle storage unit 142 includes N storage areas θ (1) to θ (N), and the estimated torque storage unit 143 includes N storage areas τ (1) to τ (N). The estimated machine rotation angle information stored in the angle storage unit 142 is θ [N], and the estimated torque information stored in the estimated torque storage unit 143 is τ [N]. The estimated machine rotation angle information θ [N] and the estimated torque information τ [N] can be handled as an array as follows.

However, i is a positive index from 1 to N, and each element of the estimated machine rotation angle information θ [N] and the estimated torque information τ [N] is associated with the index i.

θ (1) is related to τ (1), and an estimated machine rotation angle value θ ^ m1 that is a value of the estimated machine rotation angle θ ^ m is stored in the storage area θ (1), and the storage area τ (1 ) Stores an estimated torque value τ ^ m1 which is a value of the estimated torque τ ^ corresponding to the estimated mechanical rotation angle value θ ^ m1. An estimated machine rotation angle value θ ^ mi is stored in the storage area θ (i) of the index i, and an estimated torque value τ ^ mi is stored in the storage area τ (i) of the index i. The estimated mechanical rotation angle value θ ^ mN is stored in the storage area θ (N) when the index i is N, and the estimated torque value τ ^ mN is stored in the storage area τ (N) when the index i is N. Is done. That is, the following equation (33) is obtained.
θ (i) = θ ^ mi
τ (i) = τ ^ mi (33)
However, the index i is a positive number from 1 to N.

FIG. 45 shows an example in which the constant machine rotation angle value θ ^ mi and the estimated torque value τ ^ mi are stored in the angle storage unit 142 and the estimated torque storage unit 143, respectively. In FIG. 45, the estimated machine rotation angle value θ ^ m1 when the index i is 1 is the stored rotation angle value θ1, and the estimated torque value τ ^ m1 when the index i is 1 is the stored estimated torque value τ1. The estimated mechanical rotation angle value θ ^ mi of the index i is the stored rotation angle value θi, and the estimated torque value τ ^ mi of the index i is the stored estimated torque value τi.

In step S202, the estimated machine rotation angle value θ ^ mi of the estimated machine rotation angle θ ^ m of the electric motor 1 is stored in the storage area θ (i) of the angle storage unit 142 at the index i, and the estimated torque of the estimated torque τ ^ When the storage of the value τ ^ mi in the storage area τ (i) of the estimated torque storage unit 143 is completed, the process proceeds to step S203. In step S203, the index i is incremented by 1. In step S204, it is determined whether or not the index i exceeds the resolution N. If the index i is less than or equal to the resolution N, the process returns to step S202. On the other hand, if the index i exceeds the resolution N in step S204, step S102 of the recording mode execution process is terminated, and the process proceeds to step S103.

In step S103, the torque switch (TQ switch) 128 is turned off and the switching signal swt is set to 1. In step S104, the initial value of the index k is set to 0, and the process proceeds to step S105 of the output mode execution process.

As shown in FIG. 48, in step S301, the estimated machine rotation angle θ ^ m of the electric motor 1 is read, and the estimated machine rotation angle θ ^ m is read from the stored rotation angle values θi stored in the angle storage unit 142. The index i for the stored rotation angle value θi close to the value of is determined. Next, in step S302, as in step S002 of the flow chat in FIG. 38, the absolute value | Δωerr | of the rotational speed difference variation Δωerr | is calculated, and the calculated absolute value of the rotational speed difference variation Δωerr is calculated. | Δωerr | is compared with the specified threshold value ε1. If it is determined in step S302 that the absolute value | Δωerr | of the fluctuation amount Δωerr of the rotational speed difference is equal to or larger than ε1 (when the determination result is No), the process proceeds to step S303. In step S303, the index k is incremented by 1, and a new index k is generated as shown in Expression (34).
k = k + 1 (34)
However, the index k is limited to the maximum value kmax of the designated index k.

If it is determined in step S302 that the absolute value | Δωerr | of the fluctuation amount Δωerr of the rotational speed difference is smaller than ε1 (when the determination result is Yes), the process proceeds to step S304 without adjusting the index k. In step S304, torque command τ ** is output under the following conditions using the sum of index i and index k as an index of estimated torque storage unit 143.
When (i + k) ≦ N, the torque command τ ** is set to τ (i + k).
When (i + k)> N, the torque command τ ** is set to τ (i + k−N).

If the torque command τ ** is output in step S304, the output mode execution process is terminated and the process proceeds to step S106. In step S106, the state of the interface switch (IF switch) 122 is monitored. If the interface switch (IF switch) 122 is OFF (Yes in step S106), the process proceeds to step S107. On the other hand, if the interface switch (IF switch) 122 is ON (No in step S106), the process returns to step S105. In step S107, the switching signal swt is set to 0, and the output of the torque command τ ** in the learning algorithm processing unit 127 is stopped.

49 to 53, it will be described that the rotational speed pulsation of the electric motor 1 can be reduced by the operation of the learning algorithm processing unit 127. FIG. FIG. 49 is a diagram showing waveforms of load torque and output torque of the motor before learning by the learning algorithm processing unit of FIG. FIG. 50 is a diagram showing waveforms of load torque and output torque of the motor after learning by the learning algorithm processing unit of FIG. FIG. 51 is a diagram showing the rotation speed of the electric motor before and after learning by the learning algorithm processing unit of FIG. FIG. 52 is a diagram showing the FFT analysis result of the rotation speed of the motor before learning by the learning algorithm processing unit of FIG. FIG. 53 is a diagram showing an FFT analysis result of the rotation speed of the electric motor after learning by the learning algorithm processing unit of FIG.

49, FIG. 50, and FIG. 51 show the results when the motor 1 is driven at a rotational speed command value of 1638 rpm (that is, the mechanical frequency is 27.3 Hz) in the case of single operation. 49 and 50, the horizontal axis represents the machine rotation angle [rad], and the vertical axis represents the torque [Nm]. In FIG. 51, the horizontal axis represents time [s], and the vertical axis represents the rotation speed [rpm]. 52 and 53, the horizontal axis represents the frequency order, and the vertical axis represents the vibration level [rpm].

49 and 50 show the load torque and the output torque of the electric motor 1 with respect to the mechanical rotation angle of the electric motor 1 up to one mechanical angle. FIG. 49 shows the load torque waveform 130 and the output torque waveform 131 when the learning process of the learning algorithm processing unit 127 is not performed (that is, before learning). As shown in FIG. 49, it was found that the output torque waveform 131 of the electric motor 1 is delayed in time from the load torque waveform 130. FIG. 50 shows the load torque waveform 132 and the output torque waveform 133 when the learning process of the learning algorithm processing unit 127 is performed (that is, after learning). As shown in FIG. 50, it was confirmed that the output torque waveform 133 of the electric motor 1 almost overlaps with the load torque waveform 132.

FIG. 51 shows the actual rotational speed waveform 136 of the electric motor 1 before learning, during learning, and after learning. FIG. 52 shows a pulsation component amplitude spectrum 134 which is a spectrum result of FFT analysis of the actual rotational speed waveform 136 of the electric motor 1 in the period TA1 of FIG. FIG. 53 shows a pulsation component amplitude spectrum 135 which is a spectrum result of FFT analysis of the actual rotational speed waveform 136 of the electric motor 1 in the period TA2 of FIG. In FIG. 51, the amplitude of the pulsating component of the rotational speed waveform 136 of the electric motor 1 after learning is significantly reduced from the amplitude of the pulsating component before learning. Specifically, in FIGS. 52 and 53, pulsation component amplitude 1f

characteristics

137 and 138 having a frequency order of 1f are amplitudes of pulsation components at a frequency of 27.3 Hz of the 1f component. The pulsation component amplitude 1f characteristic 137 in FIG. 52 is the result before learning, and the pulsation component amplitude 1f characteristic 138 in FIG. 53 is the result after learning. The amplitude value of the pulsating component amplitude 1f characteristic 137 is 155 rpm, and the amplitude value of the pulsating component amplitude 1f characteristic 138 is 18 rpm. As described above, it was confirmed that the amplitude of the pulsating component at the frequency f1f corresponding to the frequency order 1f can be reduced from 155 rpm to 18 rpm by performing the learning process of the learning algorithm processing unit 127.

Since inverter control device 17 of the sixth embodiment includes torque command value switching unit 120 and torque learning unit 121, the output torque of motor 1 of compressor 80 substantially overlaps with the load torque, and the pulsation of the rotational speed of motor 1. Can be greatly reduced. Further, the torque learning unit 121 includes an angle storage unit 142 that stores the estimated machine rotation angle θ ^ m of the electric motor 1, and an estimated torque storage unit 143 that stores the estimated torque τ ^ corresponding to the estimated machine rotation angle θ ^ m. Is provided. The estimated torque information τ [N] stored in the estimated torque storage unit 143 is a torque pattern in the estimated torque τ ^. Since the torque learning unit 121 includes the angle storage unit 142 and the estimated torque storage unit 143, when the operating condition of the compressor 80 is changed, the learning algorithm processing unit 127 performs the learning process again so that the angle storage unit 142 and The estimated torque storage unit 143 is updated. Therefore, it is not necessary to store a large number of torque patterns in the operating conditions in advance. For this reason, the inverter control device 17 of the sixth embodiment can reduce the capacity of the storage device 302 shown in FIG.

In the conventional inverter control device, a torque pattern is prepared in advance, a reference load torque pattern is stored in advance, and the stored torque pattern is adjusted every time the operating condition of the compressor 80 is different. A device to reduce the capacity of the storage device was required. On the other hand, in the inverter control device 17 according to the sixth embodiment of the present invention, the estimated torque τ ^ that is the estimated value of the output torque automatically learned by the learning algorithm processing unit 127 of the torque learning unit 121 is actually used. The torque control unit 48 controls the load torque to approach. For this reason, even if the load torque of the compressor 80 fluctuates, the inverter control device 17 according to the sixth embodiment brings the output torque closer to the load torque without storing the load torque pattern prepared in advance. Can do. Furthermore, since the inverter control device 17 of the sixth embodiment only needs to store the automatically learned estimated torque τ ^, the capacity of the storage device (storage device 302 shown in FIG. 14) of the inverter control device 17 is reduced. can do.

Further, in the inverter control device 17 of the sixth embodiment, since the magnetic flux estimation unit 18 is provided, the mechanical rotation angle and the rotational speed of the electric motor 1 are estimated by the magnetic flux estimation unit 18 regardless of the state of the torque switch 128. The stability of the learning process in the learning algorithm processing unit 127 of the torque learning unit 121 can be improved.

It should be noted that the present invention can be freely combined with each other within the scope of the invention, and each embodiment can be appropriately modified or omitted.

DESCRIPTION OF SYMBOLS 1 ... Electric motor, 7 ... Operation mode discrimination | determination part, 8 ... Electric power estimation part, 9 ... Electric power pulsation extraction part, 12 ... Inverter output voltage control part, 13 ... Rotational speed command value generation part, 16 ... Inverter, 17 ... Inverter control apparatus , 18 ... Magnetic flux estimation part, 36 ... Torque pulsation extraction part, 46 ... Speed control part, 48 ... Torque control part, 60 ... Operation mode discrimination part, 61 ... Electric power pulsation extraction part, 80 ... Compressor (machine to be driven), 83a ... First compression unit, 83b ... Second compression unit, 84 ... Shaft (rotary shaft), 100 ... Inverter compressor, 120 ... Torque command value switching unit, 142 ... Angle storage unit, 143 ... Estimated torque storage unit, 180 ... torque pulsation extraction unit, P ^ ... estimated power, τ ^ ... estimated torque, τ * ... torque command, τ ** ... torque command (corrected torque command), τref ... torque correction command, τ ^ kf ... torque pulsation component ( Synthetic (Luc pulsation component), ω ^ e ... estimated rotational speed, ωe * ... rotational speed command, Idq * ... dq-axis current command vector (current command), Vdq * ... dq-axis voltage command vector (previous voltage command), Vuvw * ... Voltage command vector (voltage command), || xkf ||, || x2f ||, || xNf || ... Amplitude (amplitude of power pulsation), || τkf ||, || τ1f | |, | ΤNf ||: Amplitude (amplitude of torque pulsation), ωe0: Reference rotational speed command value, θ ^ m: Estimated rotational angle (estimated mechanical rotational angle), θi: Stored rotational angle value, θc: Phase amount (initial) Phase), ωerr ... rotational speed difference, τi ... stored estimated torque value, τ ^ u ... adjusted estimated torque.

Claims

An inverter control device that controls an electric motor that drives a machine to be driven based on an AC voltage converted by an inverter,
The electric motor is operated in an operation mode having a fluctuation component frequency of electric power or shaft torque in electric power fluctuation during one rotation of the mechanical angle,
The operation mode of the motor is estimated based on the frequency component related to the estimated component of the estimated power or the estimated torque and the variation component frequency of the estimated torque from the current of the power line that applies the control voltage to the motor. An operation mode determination unit for determining
A rotation speed command value generation unit that generates a rotation speed command according to the operation mode determined by the operation mode determination unit;
And an inverter output voltage control unit that generates a voltage command of the control voltage to be applied to the electric motor based on the value of the rotation speed command.
An inverter control device that controls an electric motor that drives a machine to be driven based on an AC voltage converted by an inverter,
The electric motor is operated in an operation mode having a fluctuation component frequency of electric power or shaft torque in electric power fluctuation during one rotation of the mechanical angle,
Based on the frequency component related to the fluctuation component frequency in the estimated power that is the estimated power that is the power of the motor estimated from the current of the power line that applies the control voltage to the motor or the estimated shaft torque of the motor, A rotation speed command value generation unit for generating a rotation speed command;
And an inverter output voltage control unit that generates a voltage command of the control voltage to be applied to the electric motor based on the value of the rotation speed command.
An operation mode determination unit that determines the operation mode of the electric motor based on the frequency component related to the fluctuation component frequency of the estimated power or the estimated torque;
The inverter control device according to claim 2, wherein the rotation speed command value generation unit generates the rotation speed command according to the operation mode determined by the operation mode determination unit.
The operation mode determination unit
A power estimation unit for calculating the estimated power;
A power pulsation extracting unit that extracts an amplitude of a power pulsation having a fluctuation component frequency that is a value that is an integer multiple of one or more of the mechanical rotation frequency of the electric motor in the estimated power,
The inverter control device according to claim 1, wherein the operation mode is determined based on an amplitude of the power pulsation.
The operation mode determination unit
A power estimation unit for calculating the estimated power;
A torque pulsation extraction unit that extracts the amplitude of torque pulsation having a fluctuation component frequency in the estimated torque calculated from the estimated power,
The inverter control device according to claim 1, wherein the operation mode is determined based on an amplitude of the torque pulsation.
The torque pulsation extracting unit
Determining a selection frequency which is the most dominant fluctuation component frequency in the extracted amplitude of the torque pulsation, and generating a synthesized torque pulsation component having the selection frequency and the amplitude of the torque pulsation of the selection frequency;
The inverter output voltage controller is
A speed control unit that generates a torque command based on the value of the rotational speed command;
A torque control unit that generates a current command based on the value of the torque command and the combined torque pulsation component;
The inverter control device according to claim 5, wherein the voltage command is generated based on the current command.
The inverter output voltage controller is
A motor phase estimator that estimates a machine rotation angle of the motor and generates an estimated machine rotation angle based on the estimated rotation speed that is the rotation speed of the motor estimated using the current command and the current of the power line. When,
Based on the difference between the estimated rotational speed and the rotational speed command, and the estimated mechanical rotational angle, a torque correction command that is one of the torque command and a corrected torque command that corrects the phase of the estimated torque is generated. A torque command value switching unit for performing
The current command is generated based on the torque correction command generated by the torque command value switching unit,
The inverter control device according to claim 6, wherein the voltage command is generated based on the current command.
The inverter output voltage controller is
An angle storage unit for storing the estimated machine rotation angle with a predetermined resolution;
An estimated torque storage unit that stores the estimated torque corresponding to the estimated machine rotation angle;
Based on the selected storage rotation angle value that is the storage rotation angle value of the angle storage unit selected based on the estimated machine rotation angle and the difference between the estimated rotation speed and the rotation speed command, the estimated machine rotation angle is calculated. Adjusting, selecting the adjusted new stored rotation angle value from the angle storage unit, and outputting the estimated storage torque value corresponding to the new stored rotation angle value from the estimated torque storage unit. The inverter control device according to claim 7, wherein a value is generated as the correction torque command.
The inverter output voltage controller is
A magnetic flux estimator that generates an estimated rotational speed that is a rotational speed of the electric motor estimated using the current command and the current of the power line;
The torque pulsation extracting unit
The initial phase of the torque pulsation at the selected frequency is corrected based on the difference between the estimated rotational speed and the rotational speed command, and the estimated rotational angle of the electric motor estimated based on the estimated rotational speed, and this correction A corrected composite torque pulsation component that is the generated composite torque pulsation component,
The inverter output voltage controller is
The current command is generated based on the corrected combined torque pulsation component generated by the torque pulsation extraction unit,
The inverter control device according to claim 6, wherein the voltage command is generated based on the current command.
The operation mode determination unit
A phase adjustment unit that advances the phase of the estimated torque and generates an adjusted estimated torque that is the estimated torque that has been advanced in phase;
The torque pulsation extracting unit
A composite torque pulsation component having the selected frequency and the amplitude of the torque pulsation at the selected frequency is determined by determining a selection frequency that is the most dominant fluctuation component frequency in the amplitude of the torque pulsation extracted in the adjusted estimated torque. The inverter control device according to claim 6, wherein:
11. One or more compression parts, The said electric motor which drives all the said compression parts by one rotating shaft, Control based on the alternating voltage converted from the said motor with the inverter, Any one of 3 to 10 An inverter compressor provided with the inverter control device according to Item.
The inverter control device according to claim 2, wherein one or more compression units, the electric motor that drives all of the compression units with a single rotating shaft, and an inverter control device according to claim 2 that controls the electric motor based on an alternating voltage converted by an inverter. Inverter compressor.
The operation mode determined by the operation mode determination unit is related to the number of compression operations of the compression unit,
The value of the rotational speed command in which only one compression unit is performing compression operation is set as a reference rotational speed command value,
The rotation speed command value generation unit indicates a value obtained by dividing the reference rotation speed command value by the number of compression operations according to the number of compression operations associated with the operation mode when the operation mode is changed. The inverter compressor according to claim 11, wherein the rotation speed command is generated.