WO2023209803A1

WO2023209803A1 - Device for controlling ac rotary machine

Info

Publication number: WO2023209803A1
Application number: PCT/JP2022/018875
Authority: WO
Inventors: 英明徳永; 雄也久野; 晃古川; 亮中村
Original assignee: 三菱電機株式会社
Priority date: 2022-04-26
Filing date: 2022-04-26
Publication date: 2023-11-02

Abstract

Provided is a device for controlling an AC rotary machine which makes it possible to determine demagnetization of a rotor magnet without using the impedance characteristics of the AC rotary machine. The control device (30) for an AC rotary machine uses m sets of voltage command values (Vqo), the voltage values (Iq) of m sets of armature windings, the resistance value (Ra) of the armature windings, and a rotational angular velocity (ω) as a basis to estimate interlinkage flux linked to the armature windings and determines whether demagnetization of a magnet is occurring on the basis of the result of comparing the estimated value of interlinkage flux and a demagnetization determination value (Thφ).

Description

AC rotating machine control device

This application relates to a control device for an AC rotating machine.

The motor drive device disclosed in Patent Document 1 has the following characteristics: rotational angular velocity, magnetic flux linkage by permanent magnets when no demagnetization occurs, d-axis inductance, d-axis current value, armature winding resistance value, and q-axis Based on the current value, calculate the standard q-axis voltage value when no demagnetization occurs, and compare the standard q-axis voltage value and the actual q-axis voltage value to estimate the amount of demagnetization. ing.

Japanese Patent Application Publication No. 2005-51892

The technique of Patent Document 1 requires the use of d-axis inductance. The d-axis inductance varies due to magnetic saturation. Therefore, when the setting accuracy of the d-axis inductance deteriorates, the calculation accuracy of the reference q-axis voltage value deteriorates, and the accuracy of demagnetization determination deteriorates. Furthermore, when the stator is provided with two or more sets of armature windings, the mutual inductance between the sets also needs to be set accurately. Further, when the rotor is provided with a field winding, the inductance of the field winding must also be set accurately. That is, in the technique of Patent Document 1, an erroneous determination occurs unless the inductance characteristics of the AC rotating machine are set accurately.

Therefore, an object of the present application is to provide a control device for an AC rotating machine that can determine demagnetization of a rotor magnet without using the inductance characteristics of the AC rotating machine.

The control device for an AC rotating machine according to the present application controls an AC rotating machine that has a rotor provided with magnets and a stator provided with m sets of armature windings (m is a natural number of 1 or more) to a power converter. A control device for an AC rotating machine that is controlled via a
a rotation detection unit that detects a rotational angular velocity in electrical angle of the rotor;
a voltage command value calculation unit that calculates a voltage command value for each group;
a switching control unit that applies voltage to the armature winding by turning on and off a switching element included in the power converter based on the voltage command value for each set;
Based on m sets of the voltage command values, m sets of current values of the armature windings, resistance values of the armature windings, and the rotational angular velocity, the interlinkage magnetic flux interlinking with the armature windings is determined. A flux linkage estimator for estimating,
a demagnetization determination unit that determines whether demagnetization of the magnet has occurred based on a comparison result between the estimated value of the flux linkage and the demagnetization determination value;
It is equipped with the following.

According to the control device for an AC rotating machine according to the present application, the voltage equation can be calculated using the inductance characteristics of the AC rotating machine without directly estimating the armature linkage flux due to the magnet and the armature reaction flux due to the d-axis current. It is possible to indirectly estimate armature flux linkage based on m sets of voltage command values, m sets of armature winding current values, armature winding resistance values, and rotational angular velocity. can. Therefore, the armature flux linkage can be estimated without using the inductance characteristics of the AC rotating machine. Based on the comparison result between the estimated value of the armature flux linkage and the demagnetization determination value, it is possible to determine whether demagnetization of the magnet has occurred.

1 is a schematic configuration diagram of an AC rotating machine and a control device for the AC rotating machine according to Embodiment 1. FIG. 1 is a schematic block diagram of a control device according to Embodiment 1. FIG. 1 is a hardware configuration diagram of a control device according to Embodiment 1. FIG. 5 is a time chart illustrating ringing behavior according to the first embodiment. FIG. 2 is a schematic block diagram of a control device including a winding temperature acquisition unit according to Embodiment 1. FIG. FIG. 3 is a diagram illustrating a range of variation in magnetic flux linkage when using a fixed resistance value according to the first embodiment. FIG. 3 is a diagram illustrating a variation range of magnetic flux linkage when using a variable resistance value according to the first embodiment. FIG. 3 is a diagram illustrating demagnetization determination when using one estimated value and one determination value according to the first embodiment. FIG. 3 is a diagram illustrating demagnetization determination when one estimated value and two determination values are used according to the first embodiment. FIG. 3 is a diagram illustrating demagnetization determination when two estimated values and one determination value are used according to the first embodiment. FIG. 3 is a diagram illustrating demagnetization determination when two estimated values and two determination values are used according to the first embodiment. FIG. 3 is a diagram illustrating operating points at which estimated values vary according to the first embodiment.

1. Embodiment 1
A control device 30 (hereinafter simply referred to as control device 30) for an AC rotating machine according to Embodiment 1 will be described with reference to the drawings. FIG. 1 is a schematic configuration diagram of an AC rotating machine 1, a power converter, and a control device 30 according to the present embodiment.

1-1. AC rotating machine 1
The AC rotating machine 1 includes a stator 18 and a rotor 14. The stator 18 is provided with m sets of armature windings (m is a natural number of 1 or more). In this embodiment, m=2 is set, and a first three-phase armature winding and a second set of three-phase armature windings are provided. The stator 18 includes a first set of three-phase armature windings Cu1, Cv1, and Cw1 of U1 phase, V1 phase, and W1 phase, and a second set of three-phase armature windings of U2 phase, V2 phase, and W2 phase. Child windings Cu2, Cv2, and Cw2 are provided. The three-phase armature windings of each set may be star-connected or delta-connected.

The rotor 14 is provided with magnets. In this embodiment, the rotor 14 is provided with a field winding 7. Further, the rotor 14 is also provided with permanent magnets 12 . Demagnetization of the permanent magnet 12 is determined by the demagnetization determination described later.

The rotor 14 is provided with a rotation sensor 15 that detects the rotation angle (rotation angle) of the rotor 14. The output signal of the rotation sensor 15 is input to the control device 30. As the rotation sensor 15, various sensors such as a Hall element, a resolver, or an encoder are used. The rotation sensor 15 may not be provided, and the rotation angle (magnetic pole position) may be estimated based on current information etc. obtained by superimposing a harmonic component on a current command value, which will be described later (so-called sensorless method).

1-2. Inverter A first set of inverters 4a and a second set of inverters 4b are provided as power converters. The first set of inverters 4a performs power conversion between the DC power supply 2 and the first set of three-phase armature windings. The second set of inverters 4b performs power conversion between the DC power supply 2 and the second set of three-phase armature windings.

In the first set of inverters 4a, a high potential side switching element SP1 connected to the high potential side of the DC power supply 2 and a low potential side switching element SN1 connected to the low potential side of the DC power supply 2 are connected in series. Three sets of series circuits (legs) are provided, one for each of the three phases. The connection point of the two switching elements in the series circuit of each phase is connected to the winding of the corresponding phase.

Specifically, in the U1 phase series circuit, the U1 phase high potential side switching element SPu1 and the U1 phase low potential side switching element SNu1 are connected in series, and the connection point of the two switching elements is the U1 phase switching element SPu1. It is connected to armature winding Cu1. In the V1 phase series circuit, the V1 phase high potential side switching element SPv1 and the V1 phase low potential side switching element SNv1 are connected in series, and the connection point between the two switching elements is the V1 phase armature winding Cv1. It is connected to the. In the W1 phase series circuit, the W1 high potential side switching element SPw1 and the W1 phase low potential side switching element SNw1 are connected in series, and the connection point of the two switching elements is connected to the W1 phase armature winding Cw1. It is connected.

In the second set of inverters 4b, a high potential side switching element SP2 connected to the high potential side of the DC power supply 2 and a low potential side switching element SN2 connected to the low potential side of the DC power supply 2 are connected in series. Three sets of series circuits (legs) are provided, one for each of the three phases. The connection point of the two switching elements in the series circuit of each phase is connected to the winding of the corresponding phase.

Specifically, in the U2 phase series circuit, the U2 phase high potential side switching element SPu2 and the U2 phase low potential side switching element SNu2 are connected in series, and the connection point of the two switching elements is the U2 phase switching element SPu2. It is connected to armature winding Cu2. In the V2 phase series circuit, the V2 phase high potential side switching element SPv2 and the V2 phase low potential side switching element SNv2 are connected in series, and the connection point between the two switching elements is the V2 phase armature winding Cv2. It is connected to the. In the W2 phase series circuit, the W2 high potential side switching element SPw2 and the W2 phase low potential side switching element SNw2 are connected in series, and the connection point of the two switching elements is connected to the W2 phase armature winding Cw2. It is connected.

The first set of inverters 4a and the second set of inverters 4b are connected to one DC power supply 2. One smoothing capacitor 3 is connected to the DC power supply 2 in parallel. Note that a smoothing capacitor may be provided in each of the first set and the second set of

inverters

4a and 4b.

As the switching element, an IGBT (Insulated Gate Bipolar Transistor) with diodes connected in anti-parallel, a MOSFET (Metal Oxide Semiconductor Field Effect Transistor), a bipolar transistor with diodes connected in anti-parallel, etc. are used. The gate terminal of each switching element is connected to the control device 30 via a gate drive circuit or the like. Each switching element of the first set of inverters 4a is turned on or off by a switching signal for the first set output from the control device 30. Each switching element of the second set of inverters 4b is turned on or off by a switching signal for the second set output from the control device 30.

The DC power supply 2 outputs a DC voltage Vdc to the first and second sets of

inverters

4a and 4b. The DC power source 2 may be any device that outputs a DC voltage Vdc, such as a battery, a DC-DC converter, a diode rectifier, a PWM rectifier, or the like.

A first set of armature current sensors 5a and a second set of armature current sensors 5b are provided for detecting currents flowing through the armature windings of each phase of the first and second sets. The first and second sets of armature

current sensors

5a and 5b are current sensors such as shunt resistors or Hall elements. The output signals of the first set and the second set of armature

current sensors

5a, 5b are input to the control device 30.

In this embodiment, each set of armature

current sensors

5a, 5b is provided on an electric wire that connects a series circuit of switching elements of each phase and a winding of each phase. Note that each set of armature

current sensors

5a, 5b may be connected in series to a series circuit of switching elements of each phase. Alternatively, each set of current sensors is provided on a wire connecting each set of

inverters

4a, 4b and the DC power supply 2, and the current in each phase winding of each set is may be detected.

1-3. converter 9
A converter 9 is provided as a power converter. Converter 9 has a switching element and performs power conversion between DC power supply 2 and field winding 7 . In the present embodiment, the converter 9 has a high potential side switching element SP connected to the high potential side of the DC power supply 2 and a low potential side switching element SN connected to the low potential side of the DC power supply 2 connected in series. It is an H-bridge circuit with two sets of connected series circuits. The connection point between the switching element SP1 on the high potential side and the switching element SN1 on the low potential side in the first series circuit 28 is connected to one end of the field winding 7, and the high potential in the second series circuit 29 is connected to one end of the field winding 7. A connection point between the switching element SP2 on the side and the switching element SN2 on the low potential side is connected to the other end of the field winding 7.

As the switching element of the converter 9, an IGBT with diodes connected in anti-parallel, a bipolar transistor with diodes connected in anti-parallel, a MOSFET, etc. are used. The gate terminal of each switching element is connected to the control device 30 via a gate drive circuit or the like. Therefore, each switching element is turned on or off by a switching signal output from the control device 30.

Note that the converter 9 may be replaced with another one, such as by replacing the switching element SN1 on the low potential side of the first series circuit 28 with a diode, or replacing the switching element SP2 on the high potential side of the second series circuit 29 with a diode. It may also be configured as follows.

The field current sensor 6 is a current detection circuit that detects a field current value If, which is a current flowing through the field winding 7. In this embodiment, field current sensor 6 is provided on an electric wire that connects field winding 7 and converter 9. The field current sensor 6 may be provided at another location where the field current value If can be detected. The output signal of the field current sensor 6 is input to the control device 30. The field current sensor 6 is a current sensor such as a Hall element or a shunt resistor.

1-4. Control device 30
The control device 30 controls the AC rotating machine 1 via a power converter (in this example, a first set and a second set of

inverters

4a, 4b, and a converter 9). As shown in FIG. 2, the control device 30 has functions such as a rotation detection section 31, a current detection section 32, a voltage command value calculation section 33, a switching control section 34, a magnetic flux linkage estimation section 35, and a demagnetization determination section 36. It has a department. Each function of the control device 30 is realized by a processing circuit included in the control device 30. Specifically, as shown in FIG. 3, the control device 30 includes, as a processing circuit, an arithmetic processing device 90 (computer) such as a CPU (Central Processing Unit), a storage device 91 that exchanges data with the arithmetic processing device 90, It includes an input circuit 92 that inputs external signals to the arithmetic processing device 90, an output circuit 93 that outputs signals from the arithmetic processing device 90 to the outside, and a communication circuit 94 that performs data communication with an external device.

The arithmetic processing unit 90 includes an ASIC (Application Specific Integrated Circuit), an IC (Integrated Circuit), a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array), various logic circuits, and various signal processing circuits. You can. Further, a plurality of arithmetic processing units 90 of the same type or different types may be provided, and each process may be shared and executed. As the storage device 91, a RAM (Random Access Memory) configured to be able to read and write data from the arithmetic processing unit 90, a ROM (Read Only Memory) configured to be able to read data from the arithmetic processing unit 90, etc. It is equipped. The input circuit 92 is an A/D to which various sensors such as the rotation sensor 15, each set of armature

current sensors

5a and 5b, and the field current sensor 6 are connected, and inputs the output signals of these sensors to the arithmetic processing unit 90. Equipped with converters, etc. The output circuit 93 is connected to electrical loads such as a gate drive circuit that turns on and off the switching elements of the first and second sets of

inverters

4a and 4b and the converter 9, and receives control signals from the arithmetic processing unit 90 to these electrical loads. It is equipped with a drive circuit etc. that outputs. The communication circuit 94 communicates with external devices.

Each function of each of the control units 31 to 36 provided in the control device 30 is performed by the arithmetic processing device 90 executing software (program) stored in a storage device 91 such as a ROM, and the storage device 91 and input circuit 92. , and other hardware of the control device 30 such as the output circuit 93. Note that various setting data used by each of the control units 31 to 36 and the like are stored in a storage device 91 such as a ROM as part of software (program). Each function of the control device 30 will be described in detail below.

1-4-1. Rotation detection section 31
The rotation detection unit 31 detects the magnetic pole position θ of the rotor in electrical angle (rotation angle θ of the rotor) and the rotational angular velocity ω. In the present embodiment, the rotation detection unit 31 detects the magnetic pole position θ (rotation angle θ) and rotational angular velocity ω in electrical angle based on the output signal of the rotation sensor 15. Note that the electrical angle is the angle obtained by multiplying the mechanical angle of the rotor 14 by the number of pole pairs of the magnet.

The magnetic pole position θ is set to the direction of the north pole of the magnets (in this example, electromagnets and permanent magnets) provided on the rotor. In this embodiment, the magnetic pole position θ (rotation angle θ) is the position (angle) of the magnetic pole (N pole) in electrical angle with reference to the armature winding of the U1 phase of the first set. When a phase difference is provided between the first set of armature windings and the second set of armature windings, the phase difference is taken into account and the magnetic pole position θ (rotation angle θ) for each set is determined. Calculated.

Note that the rotation detection unit 31 is configured to estimate the rotation angle (magnetic pole position) without using a rotation sensor, based on current information etc. obtained by superimposing harmonic components on the current command value. (so-called sensorless method).

1-4-2. Current detection section 32
The current detection unit 32 detects winding currents Ius1, Ivs1, and Iws1 flowing through the first set of three-phase armature windings based on the output signal of the first set of armature current sensors 5a. Further, the current detection unit 32 detects winding currents Ius2, Ivs2, and Iws2 flowing through the second set of three-phase armature windings based on the output signal of the second set of armature current sensors 5b. Note that for each set, two-phase winding currents may be detected, and the remaining one-phase winding current may be calculated based on the detected values of the two-phase winding currents.

Furthermore, the current detection unit 32 detects a field current value Ifs, which is the current flowing through the field winding 7, based on the output signal of the field current sensor 6.

1-4-3. Voltage command value calculation unit 33
The voltage command value calculation unit 33 calculates the voltage command value for each group.

The voltage command value calculation unit 33 calculates the d-axis current command value Ido and the q-axis current command value Iqo for each set using various known methods. For example, the voltage command value calculation unit 33 calculates the d-axis and q-axis current command values Ido and Iqo based on the torque command value, rotational angular velocity ω, etc. using known vector control for each set. Let the first set of d-axis and q-axis current command values be Ido1 and Iqo1, and the second set of d-axis and q-axis current command values as Ido2 and Iqo2. The torque command value may be calculated within the control device 30 or may be transmitted from an external control device. The d-axis is defined in the direction of the north pole of the magnet, and the q-axis is defined in the direction 90 electrical degrees ahead of the d-axis.

The voltage command value calculation unit 33 performs known three-phase two-phase conversion and rotational coordinate conversion on the three-phase current detection values Ius, Ivs, and Iws for each set based on the magnetic pole position θ, and converts the three-phase current detection values Ius, Ivs, and Iws into the d-axis The current detection value Ids and the q-axis current detection value Iqs are converted. Let the first set of d-axis and q-axis current detection values be Ids1 and Iqs1, and the second set of d-axis and q-axis current detection values be Ids2 and Iqs2.

The voltage command value calculation unit 33 performs known current feedback control for each set based on the d-axis and q-axis current command values Ido, Iqo, and the d-axis and q-axis current detection values Ids, Iqs, A d-axis voltage command value Vdo and a q-axis voltage command value Vqo are calculated. The first set of voltage command values for the d-axis and q-axis are Vdo1 and Vqo1, and the second set of voltage command values for the d-axis and q-axis are Vdo2 and Vqo2. Note that, for each set, the voltage command value calculation unit 33 calculates the d-axis voltage command value Vdo and the q-axis voltage command value by known feedforward control based on the d-axis and q-axis current command values Ido and Iqo. Vqo may also be calculated.

Then, the voltage command value calculation unit 33 performs known fixed coordinate transformation and two-phase three-phase transformation on the d-axis and q-axis voltage command values Vdo and Vqo for each set based on the magnetic pole position θ. Three-phase voltage command values Vuo, Vvo, and Vwo are calculated. The voltage command values for the three phases in the first set are Vuo1, Vvo1, and Vwo1, and the voltage command values for the three phases in the second set are Vuo2, Vvo2, and Vwo2. Known modulation such as space vector modulation or two-phase modulation may be applied to each set of three-phase voltage command values.

Note that the voltage command values for each set of three phases may be calculated using other known control methods such as V/f control. When V/f control is performed, a current detection value is not required, so a current sensor may not be provided.

1-4-4. Switching control section 34
The switching control unit 34 applies voltage to the armature windings by turning on and off switching elements included in each group of inverters based on the voltage command value for each group. In this embodiment, the switching control unit 34 generates, for each group, a switching signal that turns on and off a plurality of switching elements included in each group of inverters based on three-phase voltage command values Vuo, Vvo, and Vwo. . The switching control unit 34 uses known carrier comparison PWM or space vector PWM.

When carrier comparison PWM is used, the switching control unit 34 compares the carrier wave with each of the three-phase voltage command values Vuo, Vvo, and Vwo for each group, and based on the comparison results, the switching control unit 34 compares the carrier wave with each of the three-phase voltage command values Vuo, Vvo, and Vwo, Generates a switching signal to turn on and off.

When space vector PWM is used, the switching control unit 34 generates a voltage command vector from the three-phase voltage command values Vuo, Vvo, and Vwo, and calculates the seven basic voltage vectors in the PWM cycle based on the voltage command vector. The output time allocation is determined, and based on the output time allocation of the seven basic voltage vectors, a switching signal is generated to turn on and off each switching element in a PWM cycle.

<On/off control of converter 9>
The switching control unit 34 turns on and off the switching elements included in the converter 9 and applies a voltage to the field winding 7 . In the present embodiment, the switching control unit 34 sets the field current command value Ifo based on the torque command value, rotational angular velocity ω, etc., so that the field current detected value Ifs approaches the field current command value Ifo. , changes the field voltage command value Vfo, and turns on/off the plurality of switching elements of the converter 9 by PWM control based on the field voltage command value Vfo. Note that when feedforward control or control using an estimated value of the field current is performed, the field current detection value Ifs may not be used.

1-4-5. Interlinkage flux estimating unit 35 and demagnetization determining unit 36
<Principle of estimating magnetic flux linkage>
First, the principle of estimating flux linkage according to the present application will be explained. The voltage equation of the AC rotating machine 1 of this embodiment is shown in the following equation. Here, as will be described later, in order to consider the average value of the two sets of dq-axis current values and the average value of the two sets of dq-axis voltage values, the total value of the two sets of dq-axis current values, An equation using the sum of the two sets of dq-axis voltage values has been derived.

ω is the rotational angular velocity in electrical angle, Ra is the resistance value of the armature winding, s is the Laplace operator, Ld is the d-axis self-inductance, and Lq is the q-axis self-inductance. is the self-inductance of the armature windings, Md is the d-axis mutual inductance between the armature winding sets, Mq is the q-axis mutual inductance between the armature winding sets, and Lmd is the d-axis mutual inductance between the armature winding sets. It is the mutual inductance between the wire and the field winding, Rf is the resistance value of the field winding, Lf is the inductance of the field winding, and φf0 is the flux linkage due to the permanent magnet. . Note that the inductances Ld, Lq, resistance values Ra, and mutual inductances Md, Mq of each set of armature windings are the same.

In this embodiment, as shown in the following equation, the armature interlinkage magnetic flux φa interlinking with the armature winding is determined by the field winding interlinkage magnetic flux φf according to the field current value If caused by the field winding. and the interlinkage magnetic flux φf0 due to the permanent magnet.

Since the voltage that can be applied to each armature winding is less than the DC voltage Vdc, the conditions under which the dq-axis current can be stably controlled to a desired value are as follows. Therefore, by determining the field current during magnetic flux estimation control in consideration of the voltage saturation condition of equation (3), the dq-axis current can be output as the command value, and the magnetic flux can be accurately estimated.

In a steady state, the term of the Laplace operator s in equation (1) becomes 0, so equation (1) becomes as follows.

From equation (4), armature flux linkage φa can be expressed by the following equation.

Further, in the dq-axis currents caused by two or more sets of armature windings, ringing may occur even in a steady state due to interference between the currents flowing through the respective armatures. As an example, FIG. 4 shows waveforms of ringing caused by two sets of armature windings. The current detection values of the first set of armature windings and the second set of armature windings may ring in mutually opposite phases around the current command value of each set. This is due to mutual interference of armature windings between sets.

Furthermore, the current detection value may differ from one armature winding to another due to sensor error. Similarly, a difference may occur in the voltage command value calculated based on the detected current value that includes sensor error.

In this way, even if there are differences between sets in the dq-axis current values flowing through the armature windings and the dq-axis voltage command values calculated based on them, it is necessary to accurately determine demagnetization. There is.

Therefore, it is conceivable to use the average value of the two sets of dq-axis current values and the average value of the two sets of dq-axis voltage command values to calculate the armature flux linkage φa. In equations (1) to (5), equations using two sets of total values are derived in advance with this in mind.

As a result, in an AC rotating machine having two or more sets of armature windings, the armature linkage magnetic flux φa can be calculated by using the average value of the voltage command values of all sets and the average value of the current values of all sets. It is possible to make each variable used for more constant, and also to reduce the influence of sensor errors for each armature winding, so it is possible to reduce the calculation error of armature linkage magnetic flux φa. .

In this embodiment, since two sets of armature windings are provided, the average value of the q-axis voltage value Vqave and the average value of the d-axis current value Idave used for calculating the armature flux linkage φa , and the average value Iqave of the q-axis current value are given by equations (6) to (8). The armature magnetic flux linkage φa is calculated by Equation (9), which is a modification of Equation (5), using each average value.

The second term in the numerator of formula (9) is the armature reaction magnetic flux generated by the d-axis current value Id. Armature reaction flux is a variable term that varies depending on magnetic saturation characteristics. That is, in order to accurately calculate the armature reaction magnetic flux, map data showing changes in the d-axis inductance Ld and the d-axis mutual inductance Md is required, and if the accuracy of these inductances deteriorates, the equation (9 ), the calculation accuracy of the armature flux linkage φa deteriorates, and the accuracy of demagnetization determination deteriorates.

Therefore, equation (9) is modified as shown in the following equation. φd is the d-axis armature flux linkage that interlinks with the armature winding, which is the sum of the armature flux linkage φa and the armature reaction flux due to the d-axis current. That is, equation (9) has been transformed into an equation for calculating the d-axis armature flux linkage φd. In the central side of equation (10), the armature interlinkage magnetic flux φa varies due to demagnetization, and the d-axis inductance Ld and the d-axis mutual inductance Md vary due to magnetic saturation. Therefore, as shown on the right side of equation (10), based on the average value Vqave of the q-axis voltage value, the resistance value Ra of the armature winding, the average value Iqave of the q-axis current value, and the rotational angular velocity ω, Specifically, let us consider estimating the d-axis armature flux linkage φd.

Note that in a steady state, changes in the d-axis inductance Ld and the d-axis mutual inductance Md are small, so the d-axis armature linkage flux φd calculated by the right side of equation (10) can be regarded as constant. . Furthermore, since the resistance value Ra is also taken into consideration, it is possible to cope with the case where the resistance value Ra changes due to temperature change. The rotational angular velocity ω is also taken into account. Therefore, from the right side of equation (10), the d-axis armature flux linkage φd can be estimated indirectly using the voltage equation.

Furthermore, the demagnetization determination value Thφ can also be set using the right side of equation (10). That is, in a steady state at a certain operating point when no demagnetization occurs, the demagnetization determination value can be set based on the d-axis armature linkage flux φd calculated from the right side of equation (10). A method for setting the demagnetization determination value will be described later.

Unlike this embodiment, even when one set of armature windings is provided, calculation formulas for the armature flux linkage φa and the armature flux linkage φd of the d-axis are similarly derived based on the voltage equation. can. Since the mutual inductance Md on the d-axis becomes zero, by substituting Md=0 into equations (9) and (10), we obtain the calculation equation for one set of armature windings as shown in the following equation. . Therefore, even in the case of one set, the d-axis armature flux linkage φd can be estimated by the same equation as the right side of equation (10) in the case of two sets, as shown on the right side of equation (12).

Unlike this embodiment, even when three sets of armature windings are provided, armature linkage magnetic flux φa and d-axis armature linkage An arithmetic expression for the magnetic flux φd can be derived. Therefore, even in the case of three sets, the d-axis armature flux linkage φd can be estimated by the same equation as the right side of equation (10) in the case of two sets, as shown on the right side of equation (14). The same applies to 4 or more groups.

<Estimation of magnetic flux linkage and demagnetization determination>
Therefore, the flux linkage estimating unit 35 determines whether the armature windings are adjusted based on the m sets of voltage command values, the m sets of armature winding current values, the armature winding resistance values Ra, and the rotational angular velocity ω. Estimate the armature flux linkage. A detected current value or a current command value is used as the current value of each set of armature windings. The demagnetization determination unit 36 determines whether demagnetization of the magnet has occurred based on the comparison result between the estimated value of the flux linkage and the demagnetization determination value.

According to this configuration, by using the d-axis inductance, the mutual inductance between pairs, and the inductance of the field winding, the voltage Armature flux linkage can be estimated indirectly using the equation. Therefore, the armature flux linkage can be estimated without providing map data or the like to accurately calculate each inductance Ld, Md. Based on the comparison result between the estimated value of the armature flux linkage and the demagnetization determination value, it is possible to determine whether demagnetization of the magnet has occurred.

In this embodiment, demagnetization of the permanent magnet 12 provided in the rotor 14 is determined.

The demagnetization determination unit 36 transmits the determination result of the demagnetization determination to an external control device or the like. The user can determine whether or not maintenance of the AC rotating machine 1 is necessary based on the determination result of the occurrence of demagnetization. The demagnetization determination unit 36 may determine the degree of demagnetization based on the comparison result between the estimated value of the flux linkage and the demagnetization determination value. For example, the demagnetization determination unit 36 determines that the degree of demagnetization is greater as the degree of decrease in the estimated value of the flux linkage from the demagnetization determination value is greater.

In the present embodiment, the flux linkage estimation unit 35 uses the average value of the m sets of voltage command values as the m sets of voltage command values, and uses the average value of the m sets of current values as the m sets of current values. use

According to this configuration, by using the average value, it is possible to cancel the effect of ringing in the current value of each group caused by mutual interference of the armature windings between the groups, as described above, and the armature linkage magnetic flux The estimation accuracy can be improved. Further, even if an error occurs in the current sensor, the influence of the error can be reduced.

In this embodiment, the flux linkage estimation unit 35 uses the q-axis voltage command value Vqo as the voltage command value, uses the q-axis current value Iq as the current value, and calculates the estimated value of the armature flux linkage. The d-axis armature flux linkage φd interlinking with the armature winding is estimated as follows.

As shown in the following equation, the flux linkage estimation unit 35 calculates the average value Vqoave of the q-axis voltage command values Vqo1 and Vqo2 of the first and second sets, and the q-axis current values of the first and second sets. The d-axis armature linkage flux φd is estimated based on the average value Iqave of Iq1 and Iq2, the resistance value Ra of the armature winding, and the rotational angular velocity ω. The q-axis current detection value Iqs or the q-axis current command value Iqo of each group is used as the q-axis current value of each group. Note that when one set of armature windings is provided, the right side of equation (12) is used.

(1) Method for determining resistance value Ra (1-1) Estimation of φd Estimation of the d-axis armature linkage magnetic flux φd (hereinafter also simply referred to as φd) will be explained by focusing on the resistance value Ra. The resistance value Ra changes depending on the winding temperature. The method for determining demagnetization differs depending on whether the resistance value Ra is made variable or fixed according to the winding temperature. In the following, a case where the resistance value Ra is made variable and a case where the resistance value Ra is made fixed will be explained.

(1-1-1) When Ra is set to a fixed value First, regarding the case where the resistance value Ra used for estimating the d-axis armature flux linkage φd (hereinafter also simply referred to as φd) is set to a fixed value. explain. In this case, an error occurs between the actual resistance value Ra and the fixed resistance value Ra, which affects the estimation accuracy of φd.

Therefore, the fluctuation range of the resistance value is known through preliminary study, and the fixed resistance value Ra is determined by considering the tolerance and detection rate of demagnetization determination. Therefore, the maximum value Ra_max and minimum value Ra_min of the resistance value variation range are obtained in advance and reflected in the estimated value of φd.

At the same operating point, such as the rotational angular velocity ω and the q-axis current value Iq (or torque command value), the q-axis current value Iq is constant regardless of the resistance value Ra, so the second term in equation (17) is , depends on the magnitude of the resistance value Ra. Therefore, the second term of equation (17) can be expressed as Ra_max×Iq and Ra_min×Iq, respectively, using the maximum value Ra_max and minimum value Ra_min of the resistance. Ra_max×Iq and Ra_min×Iq indicate the maximum value and minimum value of Ra×Iq at each operating point.

In the following, for ease of explanation, let ω=1 and consider the numerator of formula (17). From the numerator of equation (17), the larger Vq and the smaller Ra×Iq, the larger the estimated value of φd. Therefore, by setting Ra to Ra_min, the estimated value of φd during actual operation becomes large, making it difficult to determine that demagnetization has occurred, and false positives can be reduced. Note that a false positive is a misjudgment that something that is not demagnetized is erroneously determined to be demagnetized.

On the other hand, the smaller Vq and the larger Ra×Iq, the smaller the estimated value of φd, so if you want to calculate a smaller estimated value of φd during actual operation, by setting Ra to Ra_max, The estimated value of φd during actual operation becomes smaller, making it easier to determine that demagnetization has occurred, and false negatives can be reduced. Note that a false negative is a false determination that a demagnetized object is not demagnetized.

It is also conceivable to estimate two φd using both Ra_max and Ra_min as Ra used for the estimated value of φd, and use the two estimated values of φd for the demagnetization determination. When two estimated values of φd are used, it is possible to increase the degree of freedom in determining demagnetization. The demagnetization determination method using the two estimation results of φd will be described later in sections (2) and (3).

(1-1-2) When making Ra a variable value Next, a case where the resistance value Ra used for estimating φd is variable according to the winding temperature will be described. In this case, as shown in FIG. 5, a winding temperature acquisition section 37 that detects or estimates the winding temperature of the armature winding is provided. When detecting the winding temperature, a temperature sensor is attached to the armature winding, the output signal of the temperature sensor is input to the control device 30, and the winding temperature acquisition section 37 detects the winding temperature. When estimating the winding temperature, the winding temperature acquisition unit 37 estimates the winding temperature using an estimation model such as a thermal model based on the operating state such as the current value. Then, the flux linkage estimation unit 35 refers to characteristic data in which the relationship between the winding temperature and the resistance value Ra is set in advance, and estimates the resistance value Ra based on the detected or estimated winding temperature. Then, the flux linkage estimation unit 35 estimates φd using the estimated resistance value Ra. Therefore, by using a highly accurate resistance value Ra, assuming that there is no error in each sensor detection value, it is possible to estimate φd with higher accuracy than when using a fixed resistance value Ra.

(1-2) Setting of Demagnetization Judgment Value Thφ Next, a method of setting the demagnetization judgment value Thφ will be explained. Similarly to the estimated value of φd, calculation of the d-axis armature linkage magnetic flux φdth (hereinafter also referred to as φdth for determination value setting) for setting the determination value is performed when the resistance value Ra is a fixed value or a variable value. I will explain about it.

Similarly to the estimation of φd, the setting accuracy of the demagnetization determination value Thφ becomes higher when the resistance value Ra is made variable. Therefore, if the winding temperature acquisition section 37 is provided to acquire the winding temperature, it is better to use the variable resistance value Ra also for setting the demagnetization determination value Thφ.

(1-2-1) When setting Ra to a fixed value Figure 6 shows the graph of φd at an operating point with rotational angular velocity ω and q-axis current value Iq (or torque command value) when no demagnetization occurs. Each parameter used for estimation is illustrated. Ra_max×Iq and Ra_min×Iq, and Vqo_max and Vqo_min are the maximum and minimum values of the fluctuation range of Ra×Iq at a certain operating point and the q-axis voltage command value Vqo obtained in advance verification. It shows the maximum and minimum values of the fluctuation range. The maximum value φd_max_ass of the estimated value of φd and the minimum value φd_min_ass of the estimated value of φd based on a combination of these four values are as shown in the following equation. Here, as described above, for ease of explanation, ω is set to 1.

However, if there is no error in each sensor detection value and there is no demagnetization, Ra and Vqo at a certain operating point are in a proportional relationship, and as Ra increases, Vq increases. Therefore, when Ra×Iq is maximum, Vq is also maximum, and conversely, when Ra×Iq is minimum, Vq is also minimum. Therefore, if the estimated value of φd when Ra_min is used to estimate φd is φd_min, and the estimated value of φd when Ra_max is used to estimate φd is φd_max, then the following relationship holds true.

Therefore, regardless of whether Ra_min or Ra_max is used to estimate φd, the estimated value of φd when no demagnetization occurs falls within the range from φd_min_ass to φd_max_ass, as shown in the following equation. Therefore, if the estimated value of φd falls outside the range from φd_min_ass to φd_max_ass, it can be determined that demagnetization has occurred. For example, the demagnetization determination unit 36 can determine that demagnetization has occurred when the estimated value of φd is smaller than the demagnetization determination value Thφ set in advance to φd_min_ass. In this way, as one index, φd_max_ass and φd_min_ass can be used as the demagnetization determination value Thφ.

(1-2-2) Case where Ra is made a variable value Consider the case where the resistance value Ra is made variable according to the winding temperature acquired by the winding temperature acquisition section 37 as described above. As mentioned above, if there is no error in the detection value of each sensor and there is no demagnetization, Ra and Vqo at a certain operating point are in a proportional relationship, and when Ra × Iq is maximum, Vqo is also maximum, and vice versa. When Ra×Iq is minimum, Vqo is also minimum.

FIG. 7 shows the case where φd is estimated using the estimated value of the resistance value Ra that is changed according to the winding temperature. When there is no error in each sensor detection value, the estimated value of φd fluctuates little with respect to a change in the resistance value Ra. Although shown as the maximum value φd_maxerr and the minimum value φd_minerr of the estimated value of φd due to the error of each sensor detection value when demagnetization does not occur, the fluctuation range is small. Therefore, the fluctuation range is smaller than the maximum value φd_max and the minimum value φd_min of the estimated value of φd when a fixed resistance value Ra is used. The demagnetization determination value Thφ may be set in advance to a value smaller than the minimum value φd_minerr of the estimated value of φd taking into account the error of each sensor detection value. That is, the demagnetization determination unit 36 determines that demagnetization has occurred when the estimated value of φd is smaller than the demagnetization determination value Thφ, which is preset to a value smaller than φd_minerr. In this way, by estimating φd using the variable resistance value Ra, it is possible to improve the accuracy of demagnetization determination. Furthermore, it is possible to set a demagnetization determination value Thφ that can determine minute demagnetization.

(2) Number of estimated values of φd and number of Thφ In the above, a method for making a demagnetization determination using one estimated value of φd and one demagnetization determination value Thφ has been described. In the following, a method of making a demagnetization determination using one or both of the estimated value of φd and the demagnetization determination value Thφ will be described.

(2-1) Number of estimated values of φd As explained in (1-2-2), when using variable resistance value Ra depending on the winding temperature, the estimation accuracy of φd is high, so multiple There is less need to estimate φd. Below, a method for estimating a plurality of φd when using a fixed resistance value Ra will be described.

As explained in (1-1-1), when using a fixed resistance value Ra, it is possible to estimate the two φd_max and φd_min using the variable range Ra_min and Ra_max as Ra. When estimating φd_max using Ra_min, φd_max becomes large, so false positives can be reduced. On the other hand, when estimating φd_min using Ra_max, φd_min becomes smaller, so false negatives can be reduced.

(2-2) Number of Thφ It is also possible to set a plurality of demagnetization determination values Thφ corresponding to φd_max corresponding to Ra_min and φd_min corresponding to Ra_max.

(3) Number of estimated values of φd, number of Thφ, and combination method Set the number of estimated values of φd, number of demagnetization judgment values Thφ, and combination method in consideration of demagnetization judgment tolerance and detection rate. Therefore, a plurality of demagnetization determination methods can be considered.

(3-1) When using one estimated value of φd and one Thφ FIG. 8 shows demagnetization determination when using one estimated value of φd and one demagnetization determination value Thφ.

<Pattern 1> The demagnetization determination unit 36 determines that demagnetization has not occurred when the estimated value of φd is larger than the demagnetization determination value Thφ.
<Pattern 2> The demagnetization determination unit 36 determines that demagnetization has occurred when the estimated value of φd is smaller than the demagnetization determination value Thφ.

As explained below, the resistance value Ra used to estimate φd may be a fixed value or a variable value, and the resistance value Ra used to set the demagnetization judgment value Thφ may be a fixed value or a variable value. There are cases where

(3-1-1) When Ra for estimating φd is a fixed value and Ra for setting Thφ is a fixed value For example, the fixed resistance value Ra used for estimating φd is within the variation range of resistance value. It is preset to an arbitrary value within the range of Ra_min to Ra_max. The demagnetization judgment value Thφ is set in advance to a value that takes into consideration the fluctuation range of the estimated value of φd due to resistance value fluctuations and errors in each sensor detection value, and is set to a value that will appropriately prevent the occurrence of false positives and false negatives. be done.

The demagnetization determination value Thφ is set in advance at each operating point. For example, it is the operating point of the rotational angular velocity ω and the q-axis current command value Iqo. A demagnetization determination value Thφ is preset at each operating point of the rotational angular velocity ω and the q-axis current command value Iqo. For example, the demagnetization determination unit 36 refers to map data in which the relationship between the rotational angular velocity ω, the q-axis current command value Iqo, and the demagnetization determination value Thφ is set in advance, and determines the current rotational angular velocity ω and the current q-axis current command value Iqo. A demagnetization determination value Thφ corresponding to the current value Iq (in this example, the average value Iqave of the q-axis current value) is calculated.

Alternatively, using the following equation, the demagnetization determination unit 36 determines the rotational angular velocity ω, the q-axis current value Iq (in this example, Iqave), the q-axis voltage command value Vqoth for setting the determination value, and the q-axis voltage command value Vqoth for setting the determination value. Based on the fixed resistance value Rath, calculate the d-axis armature linkage flux φdth for setting the judgment value, and subtract the offset value Δφ from the d-axis armature linkage flux φdth for setting the judgment value. , the demagnetization determination value Thφ may be set.

Here, Vqoth is a q-axis voltage command value for setting a determination value, and the q-axis voltage command value Vqo when no demagnetization occurs is set in advance. The q-axis voltage command value Vqoth for setting the determination value is set in advance at each operating point. For example, the demagnetization determination unit 36 refers to map data in which the relationship between the rotational angular velocity ω, the q-axis current value Iq, and the q-axis voltage command value Vqoth for setting the determination value is set in advance, and determines the current rotational angular velocity. A q-axis voltage command value Vqoth for determination value setting corresponding to the ω and q-axis current values Iq is calculated. The offset value Δφ and the fixed resistance value Rath are set so that the occurrence of false positives and false negatives is appropriate, taking into account the fluctuation range of the estimated value of φd due to fluctuations in resistance value and errors in each sensor detection value. Set in advance.

Alternatively, the demagnetization determination value Thφ may be set to a value corresponding to the above-mentioned φd_min_ass. In this case, the demagnetization determination unit 36 uses the preset minimum value Vqo_min of the variation range of the q-axis voltage command value when demagnetization does not occur as the q-axis voltage command value Vqoth for setting the determination value. , the demagnetization judgment value Thφ is set using the maximum value Ra_max of a preset resistance value variation range as the fixed resistance value Rath for setting the judgment value. The minimum value Vqo_min of the variation range of the q-axis voltage command value when demagnetization does not occur is preset at each operating point (rotation angular velocity ω and q-axis current value Iq). In this case, the demagnetization determination value Thφ is set using equations (23) and (24), and the offset value Δφ is set to 0.

Alternatively, the demagnetization determination value Thφ may be set to a value corresponding to the above-mentioned φd_max_ass. In this case, the demagnetization determination unit 36 uses the preset maximum value Vqo_max of the variation range of the q-axis voltage command value when demagnetization does not occur as the q-axis voltage command value Vqoth for setting the determination value. , the demagnetization judgment value Thφ is set using the minimum value Ra_min of a preset resistance value variation range as the fixed resistance value Rath for setting the judgment value. The maximum value Vqo_max of the variation range of the q-axis voltage command value when no demagnetization occurs is set in advance at each operating point (rotation angular velocity ω and q-axis current value Iq). In this case, the demagnetization determination value Thφ is set using equations (23) and (24), and the offset value Δφ is set to 0.

(3-1-2) When Ra for estimating φd is a fixed value and Ra for setting Thφ is a variable value In this case, by using a variable resistance value Ra for setting the demagnetization judgment value Thφ. , it is possible to prevent the demagnetization determination value Thφ from changing with respect to changes in the resistance value. However, since a fixed resistance value Ra is used to estimate φd, the estimated value of φd fluctuates due to fluctuations in the resistance value and errors in the detected values of each sensor. Therefore, as in (3-1-1), it is necessary to set the demagnetization determination value Thφ in consideration of the fluctuation range of the estimated value of φd, so there is not much advantage in using the variable resistance value Ra. Therefore, the explanation will be omitted.

(3-1-3) When Ra for estimating φd is a variable value and Ra for setting Thφ is a fixed value. In this case, as mentioned above, the fluctuation in the estimated value of φd due to fluctuations in the resistance value This makes it possible to improve the estimation accuracy of φd. Therefore, the demagnetization determination value Thφ is set in advance to a value that takes into account the fluctuation range of the estimated value of φd due to the error of each sensor detection value, and makes it appropriate to prevent the occurrence of false positives and false negatives. For example, the demagnetization determination value Thφ may be set in advance to a value smaller than the minimum value φd_minerr of the estimated value of φd taking into account the error of each sensor detection value. By estimating φd using the variable resistance value Ra, it is possible to improve the accuracy of demagnetization determination. Furthermore, it is possible to set a demagnetization determination value Thφ that can determine minute demagnetization. Similar to (3-1-1), the demagnetization determination value Thφ is set in advance at each operating point.

(3-1-4) When Ra for estimating φd is a variable value and Ra for setting Thφ is a variable value In this case, for example, using the following equation, the demagnetization determination unit 36 determines the rotational angular velocity. ω and q-axis current value Iq (in this example, Iqave), q-axis voltage command value Vqoth for determination value setting, and resistance value Ra estimated according to the winding temperature acquired by the winding temperature acquisition unit 37 Based on this, calculate the d-axis armature linkage magnetic flux φdth for determination value setting, and subtract the offset value Δφ from the d-axis armature linkage magnetic flux φdth for determination value setting to obtain the demagnetization determination value Thφ. Set.

Here, Vqoth is a q-axis voltage command value for setting a determination value, and the q-axis voltage command value Vqo when no demagnetization occurs is set in advance. Similar to equation (23) in (3-1-1), the q-axis voltage command value Vqoth for setting the determination value is set in advance at each operating point. The offset value Δφ is set so that the estimated value of φd and the demagnetization determination value Thφ are not too close when no demagnetization occurs. Further, the offset value Δφ is set in advance to a value that appropriately prevents the occurrence of false positives and false negatives, taking into consideration the fluctuation range of the estimated value of φd due to the error of each sensor detection value.

(3-2) When one estimated value of φd and two Thφ are used FIG. 9 shows demagnetization determination when one estimated value of φd and two demagnetization determination values Thφ are used.

The demagnetization determination unit 36 uses a larger demagnetization determination value ThφH and a smaller demagnetization determination value ThφL that is smaller than the larger demagnetization determination value ThφH. By using the two demagnetization determination values ThφH and ThφL, it is possible to individually adjust the occurrence of false positives and false negatives.

<Pattern 1> The demagnetization determination unit 36 determines that demagnetization has not occurred when the estimated value of φd is larger than the larger demagnetization determination value ThφH.
<Pattern 2> The demagnetization determination unit 36 determines that demagnetization has occurred when the estimated value of φd falls between the smaller demagnetization determination value ThφL and the larger demagnetization determination value ThφH. Alternatively, it may be determined that demagnetization has not occurred. For example, if the occurrence of false positives is suppressed, it is determined that demagnetization has not occurred. If the occurrence of false negatives is suppressed, it is determined that demagnetization has occurred.
<Pattern 3> The demagnetization determination unit 36 determines that demagnetization has occurred when the estimated value of φd is smaller than the smaller demagnetization determination value ThφL.

(3-2-1) When Ra for estimating φd is a fixed value and Ra for setting ThφH and ThφL is a fixed value Points different from (3-1-1) in the case of one demagnetization judgment value By providing two demagnetization determination values, the demagnetization determination value can be changed in accordance with the demagnetization detection level, and the degree of freedom in demagnetization determination can be increased.

For example, the smaller demagnetization determination value ThφL is set corresponding to the maximum value Ra_max of the resistance value variation range. For example, the demagnetization determination unit 36 sets the smaller demagnetization determination value ThφL using the following equation. The demagnetization determination unit 36 adds the offset value ΔφH to the smaller demagnetization determination value ThφL to set the larger demagnetization determination value ThφH. Here, the offset value ΔφH is set in advance to a value that appropriately prevents the occurrence of false positives and false negatives, taking into account the fluctuation range of the estimated value of φd due to the error of each sensor detection value. The other parameters are the same as those described above, so their explanation will be omitted.

Alternatively, the larger demagnetization determination value ThφH is set corresponding to the minimum value Ra_min of the resistance value variation range. For example, the demagnetization determination unit 36 sets the larger demagnetization determination value ThφH using the following equation. The demagnetization determination unit 36 subtracts the offset value ΔφL from the larger demagnetization determination value ThφH to set the smaller demagnetization determination value ThφL. Here, the offset value ΔφL is set in advance to a value that appropriately prevents the occurrence of false positives and false negatives, taking into account the range of variation in the estimated value of φd due to errors in each sensor detection value. The other parameters are the same as those described above, so their explanation will be omitted.

Alternatively, the smaller demagnetization judgment value ThφL is set to correspond to the maximum value Ra_max of the resistance value variation range, and the larger demagnetization judgment value ThφH is set to correspond to the minimum value Ra_min of the resistance value variation range. is set. For example, equation (27) and equation (29) are used.

(3-2-2) When Ra for estimating φd is a variable value and Ra for setting ThφH and ThφL is a fixed value In this case, as explained in (3-1-3), the resistance value The fluctuation of the estimated value of φd due to the fluctuation of φd is reduced, and the estimation accuracy of φd can be improved. Therefore, ThφH and ThφL are preset to values that will appropriately prevent the occurrence of false positives and false negatives, taking into account the range of variation in the estimated value of φd due to the error of each sensor detection value. For example, ThφL may be set to correspond to the minimum value φd_minerr of the estimated value of φd considering the error of each sensor detection value. ThφH may be set in accordance with the maximum value φd_maxerr of the estimated value of φd considering the error of each sensor detection value. Similarly to (3-1-1), ThφH and ThφL are set in advance at each operating point.

(3-2-3) When Ra for estimating φd is a variable value and Ra for setting ThφH and ThφL is a variable value In this case, for example, equation (25) in (3-1-4) Using the same following equation, the demagnetization determination section 36 determines the d-axis armature chain for determination value setting based on the resistance value Ra estimated according to the winding temperature acquired by the winding temperature acquisition section 37. Calculate the alternating magnetic flux φdth, subtract the offset value ΔφL from the d-axis armature linkage magnetic flux φdth for setting the judgment value, set ThφL, and set it to the d-axis armature linkage magnetic flux φdth for the judgment value setting. ThφH is set by adding the offset value ΔφH.

Each parameter in equation (31) is the same as equation (25), so the explanation will be omitted. The offset value ΔφL is set in consideration of the fluctuation range of the estimated value of φd due to the error of each sensor detection value and the occurrence of false negatives. The offset value ΔφH is set in consideration of the fluctuation range of the estimated value of φd due to the error of each sensor detection value and the occurrence of false positives. By adjusting ΔφL and ΔφH, the occurrence of false positives and false negatives can be adjusted individually. ΔφL may be set to zero, or ΔφH may be set to zero.

(3-3) When two estimated values of φd and one Thφ are used FIG. 10 shows demagnetization determination when two estimated values of φd and one demagnetization determination value Thφ are used. Here, the region in which demagnetization may not occur is an error region that mainly occurs between the actual resistance value and the fixed resistance value, and is the fluctuation range of the estimated value of φd when demagnetization does not occur. The demagnetization possibility region is a variation range of the estimated value of φd that may change depending on the degree of demagnetization.

The flux linkage estimation unit 35 estimates an estimated value φdH of the armature flux linkage on the d-axis on the larger side, and an estimated value φdL for the armature flux linkage on the d-axis on the smaller side.

<Pattern 1> The demagnetization determination unit 36 determines that demagnetization has not occurred when the estimated value φdL of the smaller interlinkage magnetic flux is larger than the demagnetization determination value Thφ.
<Pattern 2> The demagnetization determination unit 36 determines that demagnetization has occurred when the demagnetization determination value Thφ is between the estimated value φdL of the smaller magnetic flux linkage and the estimated value φdH of the larger magnetic flux linkage. It may be determined that demagnetization has occurred, or it may be determined that demagnetization has not occurred. For example, if the occurrence of false positives is suppressed, it is determined that demagnetization has not occurred. If the occurrence of false negatives is suppressed, it is determined that demagnetization has occurred.
<Pattern 3> The demagnetization determination unit 36 determines that demagnetization has occurred when the estimated value φdH of the larger interlinkage magnetic flux is smaller than the demagnetization determination value Thφ.

(3-3-1) When Ra for estimating φdH and φdL is a fixed value, and Ra for setting Thφ is a fixed value. For example, using the following equation corresponding to equation (17), the flux linkage estimator 35 estimates the estimated value φdH of the armature flux linkage on the larger side d-axis using the minimum value Ra_min of the preset resistance value variation range.

Using the following equation corresponding to equation (17), the flux linkage estimating unit 35 estimates the armature flux linkage of the smaller d-axis using the maximum value Ra_max of the preset resistance value variation range. Estimate the value φdL.

According to this configuration, if φdH estimated using the minimum value Ra_min of the resistance value variation range is less than Thφ, it can be accurately determined that demagnetization has occurred, taking into account the resistance value variation. If φdL estimated using the maximum value Ra_max of the resistance value fluctuation range exceeds Thφ, it can be accurately determined that demagnetization has not occurred, taking into account the resistance value fluctuation. The demagnetization determination value Thφ is set in advance to a value that appropriately prevents the occurrence of false positives and false negatives, taking into account the fluctuation range of the estimated value of φd due to the error of each sensor detection value. Similar to (3-1-1), the demagnetization determination value Thφ is set in advance at each operating point.

Alternatively, if the demagnetization judgment value Thφ is set corresponding to the maximum value Ra_max (φd_min) or φd_minerr of the resistance value fluctuation range, φdH estimated using Ra_min must be less than the demagnetization judgment value Thφ. For example, it can be determined that demagnetization has occurred reliably, and the occurrence of false negatives can be suppressed.

Alternatively, if the demagnetization judgment value Thφ is set corresponding to the minimum value Ra_min (φd_max) or φd_maxerr of the resistance value variation range, φdL estimated using Ra_max may exceed the demagnetization judgment value Thφ. For example, it can be reliably determined that there is no demagnetization, and the occurrence of false positives can be suppressed.

In addition, in the area of <Pattern 2> where it is not possible to determine whether or not demagnetization has occurred, in order to suppress the occurrence of false positives as much as possible, it is sufficient to perform demagnetization judgment by comparing only φdH with Thφ, thereby reducing the occurrence of false negatives. If suppressed as much as possible, only φdL may be compared with Thφ to determine demagnetization.

(3-3-2) When Ra for estimating φdH and φdL is a variable value, and Ra for setting Thφ is a fixed value If Ra for estimation is a variable value, the accuracy of the estimated value of φd is high. Since there is no need to provide a plurality of estimated values of φd, the explanation will be omitted.

(3-3-3) When Ra for estimating φdH and φdL is a variable value, and Ra for setting Thφ is a variable value Similarly, when Ra for estimating is a variable value, the estimated value of φd Since the accuracy is increased and there is no need to provide a plurality of estimated values of φd, a description thereof will be omitted.

(3-4) When two estimated values of φd and two Thφ are used FIG. 11 shows demagnetization determination when two estimated values of φd and two demagnetization determination values Thφ are used. The flux linkage estimating unit 35 estimates an estimated value φdH of the armature flux linkage on the d-axis on the larger side and an estimated value φdL of the armature flux linkage on the d-axis on the smaller side. Further, the demagnetization determination unit 36 uses a larger demagnetization determination value ThφH and a smaller demagnetization determination value ThφL.

<Pattern 1> The demagnetization determining unit 36 determines whether the demagnetization is performed when the estimated value φdH of the larger magnetic flux linkage and the estimated value φdL of the smaller magnetic flux linkage are larger than the demagnetization judgment value ThφH of the larger side. It is determined that this has not occurred.

<Pattern 2> The estimated value φdH of the larger side magnetic flux linkage is larger than the larger side demagnetization judgment value ThφH, and the estimated value φdL of the smaller side magnetic flux linkage is the estimation of the smaller side magnetic flux linkage. When the value is between the value φdL and the estimated value φdH of the larger interlinkage flux, the demagnetization determination unit 36 determines that demagnetization has occurred when the larger demagnetization determination value ThφH is used for the demagnetization determination. Alternatively, if the demagnetization determination unit 36 uses the smaller demagnetization determination value ThφL for the demagnetization determination, the demagnetization determination unit 36 may determine that demagnetization has not occurred. It is determined that this has not occurred.

<Pattern 3> The estimated value φdH of the larger side magnetic flux linkage is larger than the larger side demagnetization judgment value ThφH, and the estimated value φdL of the smaller side magnetic flux linkage is the estimation of the smaller side magnetic flux linkage. When the value is smaller than the value φdL, the demagnetization determination unit 36 may determine that demagnetization has occurred, or may determine that demagnetization has occurred, when using the larger demagnetization determination value ThφH for demagnetization determination. Alternatively, if the demagnetization determination unit 36 uses the smaller demagnetization determination value ThφL for demagnetization determination, it may determine that demagnetization has occurred, or the demagnetization determination unit 36 may determine that demagnetization has occurred. It may be determined that this has not occurred.

<Pattern 4> The estimated value φdH of the larger side magnetic flux linkage and the estimated value φdL of the smaller side magnetic flux linkage are the same as the estimated value φdL of the smaller side magnetic flux linkage and the estimated value φdH of the larger side magnetic flux linkage. In the case where the demagnetization determination unit 36 uses the larger demagnetization determination value ThφH for the demagnetization determination, the demagnetization determination unit 36 determines that demagnetization has occurred, or the demagnetization determination unit 36 When the smaller demagnetization determination value ThφL is used for demagnetization determination, it is determined that demagnetization has not occurred.

<Pattern 5> The estimated value of magnetic flux linkage on the larger side φdH is between the estimated value of magnetic flux linkage on the smaller side φdL and the estimated value of magnetic flux linkage on the larger side φdH, and When the estimated value φdL of the magnetic flux is smaller than the estimated value φdL of the smaller interlinkage magnetic flux, the demagnetization determination unit 36 determines whether the demagnetization is Alternatively, if the demagnetization determination unit 36 uses the smaller demagnetization determination value ThφL for demagnetization determination, it may determine that demagnetization has occurred, or it may determine that demagnetization has occurred. It may be determined that the

<Pattern 6> The demagnetization determination unit 36 determines whether the larger side flux linkage φdH and the smaller side estimated value φdL are smaller than the smaller side demagnetization determination value ThφL. It is determined that this has occurred.

(3-4-1) When Ra for estimating φdH and φdL is a fixed value, and Ra for setting ThφH and ThφL is a fixed value. Similar to (3-3-1), using equation (34) The magnetic flux linkage estimating unit 35 estimates the estimated value φdH of the armature magnetic flux linkage on the larger d-axis using the preset minimum value Ra_min of the resistance value variation range. Further, using equation (35), the flux linkage estimating unit 35 calculates the estimated value φdL of the armature flux linkage on the smaller d-axis using the maximum value Ra_max of the preset resistance value variation range. presume.

Similarly to (3-2-1), using equations (27) and (28), the smaller demagnetization judgment value ThφL is set corresponding to the maximum value Ra_max of the resistance value variation range, The larger demagnetization determination value ThφH is set by adding an offset value ΔφH to the smaller demagnetization determination value ThφL.

When setting ThφH and ThφL in this way, in <Pattern 3> and <Pattern 5>, which were undefined, the demagnetization determination unit 36 determines that the estimated value φdL of the smaller side flux linkage is If it is smaller than the magnetic determination value ThφL, it can be uniquely determined that demagnetization has occurred with certainty.

Alternatively, similarly to (3-2-1), using equations (29) and (30), the larger demagnetization judgment value ThφH is set corresponding to the minimum value Ra_min of the resistance value fluctuation range. The smaller demagnetization determination value ThφL may be set by subtracting the offset value ΔφL from the larger demagnetization determination value ThφH.

When setting ThφH and ThφL in this way, in <Pattern 2> and <Pattern 3> where the determination result was indeterminate, the demagnetization determination unit 36 determines that the estimated value φdH of the larger side flux linkage is larger. When the value is larger than the demagnetization determination value ThφH on the side, it can be uniquely determined that demagnetization has not occurred.

Alternatively, similarly to (3-2-1), using equations (27) and (28), the smaller demagnetization judgment value ThφL is set corresponding to the maximum value Ra_max of the resistance value fluctuation range. Then, using equations (29) and (30), the larger demagnetization determination value ThφH may be set corresponding to the minimum value Ra_min of the resistance value variation range.

When setting ThφH and ThφL in this way, in <Pattern 3> and <Pattern 5> where the judgment result was indeterminate, the estimated value φdL of the smaller interlinkage flux is the smaller demagnetization judgment value ThφL. If it is less than , it can be uniquely determined that demagnetization has definitely occurred. In addition, in <Pattern 2> and <Pattern 3> for which the judgment result was indeterminate, if the estimated value φdH of the larger interlinkage flux exceeds the larger demagnetization judgment value ThφH, It can be uniquely determined that no magnetism is generated.

<Correction of Thφ by rotational angular velocity ω>
As explained using Equation (17), Equation (23), etc., the estimated value of φd is changed in inverse proportion to the rotational angular velocity ω, and the demagnetization judgment value Thφ is changed in inverse proportion to the rotational angular velocity ω. . Therefore, in a region where the rotational angular velocity ω is low, even a slight variation in the rotational angular velocity ω causes a large variation in the estimated value of φd and the demagnetization determination value Thφ. Therefore, in the low rotation region, unexpected estimated values of φd and demagnetization determination value Thφ may be calculated, and it may be erroneously determined that demagnetization has occurred. Therefore, the demagnetization determination unit 36 may correct the demagnetization determination value Thφ based on the rotational angular velocity ω so that it is not easily determined that demagnetization has occurred. The demagnetization determination unit 36 decreases the demagnetization determination value Thφ when the rotational angular velocity ω is in a preset low rotation range. For example, the demagnetization determination unit 36 sets a value obtained by multiplying each demagnetization determination value Thφ calculated by equations (23), (24), etc. by a correction coefficient as the corrected demagnetization determination value Thφ. do. The correction coefficient is set to less than 1 when the rotation angular velocity ω is in the low rotation region, and is set to 1 when the rotation angular velocity ω is neither in the low rotation region nor in the high rotation region.

On the other hand, in the high rotation range where the rotational angular velocity ω is high, the maximum output torque is low, so the q-axis current value Iq tends to be small, the estimated value of φd and the demagnetization judgment value Thφ are small, and the amount of variation thereof is Often smaller. In order to reduce false negatives in such a high rotation region, the demagnetization determination unit 36 increases the demagnetization determination value Thφ when the rotational angular velocity ω is in a preset high rotation region. For example, the demagnetization determination unit 36 sets a value obtained by multiplying each demagnetization determination value Thφ calculated by equations (23), (24), etc. by a correction coefficient as the corrected demagnetization determination value Thφ. do. When the rotation angular velocity ω is in a high rotation region, the correction coefficient is set to be larger than 1, and when the rotation angular velocity ω is not in a high rotation region or a low rotation region, the correction coefficient is set to 1. Note that either the correction in the low rotation region or the correction in the high rotation region may be performed.

<Switching between execution and non-execution of demagnetization judgment according to the operating point>
The demagnetization determination unit 36 determines whether or not to perform the demagnetization determination based on the rotational angular velocity ω, the voltage command value, and the current value, and when it is determined that the demagnetization determination is to be performed, demagnetization has occurred. If it is determined that demagnetization determination is not to be performed, it is not determined whether demagnetization has occurred.

As shown in FIG. 12, for example, at a certain operating point, the estimated value of φd may vary periodically. At this operating point, the estimated value of φd may not be constant and may oscillate around the demagnetization determination value Thφ. Since the accuracy of demagnetization determination decreases at such an operating point, it is better not to perform demagnetization determination. Therefore, by setting in advance the operating point at which the accuracy of demagnetization determination decreases as the operating point at which demagnetization determination is not performed, the accuracy of demagnetization determination can be improved. At the operating point where the estimated value of φd becomes large, the difference from the demagnetization determination value Thφ increases, so the possibility of erroneous determination decreases and the accuracy of demagnetization determination improves. Therefore, the accuracy of demagnetization determination can be improved by setting in advance the operating point at which the accuracy of demagnetization determination is improved as the operating point at which demagnetization determination is executed.

<Change in demagnetization judgment value Thφ due to field current value If>
The demagnetization determination section 36 may change the demagnetization determination value Thφ based on the field current value If. In this embodiment, as shown in equation (2), the armature linkage magnetic flux φa changes according to the field current value If, and as shown in equation (10), the armature linkage flux φa of the d-axis changes according to the field current value If. The magnetic flux φd changes according to the armature linkage magnetic flux φa, which changes according to the field current value If.

As can be seen from equation (10), when the field current value If changes and the d-axis armature linkage flux φd changes, the d-axis voltage command value Vqo and the d-axis current value Iq automatically change. Therefore, the estimated value of φd and the demagnetization determination value Thφ also change automatically according to changes in the d-axis voltage command value Vqo and the d-axis current value Iq. Therefore, basically, even if the field current value If changes, the accuracy of demagnetization determination can be maintained.

Specifically, in this embodiment, the q-axis current command value Iqo is set based on the rotational angular velocity ω and the torque command value, and the field current command value Ifo is set based on the rotational angular velocity ω and the torque command value. Set. Therefore, the q-axis current value Iq, the field current value If, and the rotational angular velocity ω uniquely correspond to each other, and the q-axis voltage command value Vqo also depends on the q-axis current value Iq and the field current value If. and the rotational angular velocity ω. Therefore, the q-axis voltage command value Vqoth for judgment value setting, which is preset for each operating point of the rotational angular velocity ω and the q-axis current value Iq (or torque command value), also depends on the q-axis current value Iq and the field. The current value If and the rotational angular velocity ω uniquely correspond to each other. That is, the field current value If and the q-axis voltage command value Vqoth for setting the determination value uniquely correspond to each other via the reference parameters of the rotational angular velocity ω and the q-axis current value Iq (or torque command value). Therefore, even if the field current value If changes, the accuracy of demagnetization determination can be maintained.

However, unlike this embodiment, if the field current value If and the q-axis voltage command value Vqoth for setting the determination value do not uniquely correspond to each other via the reference parameter, in order to maintain the determination accuracy, It is necessary to change the demagnetization determination value Thφ in accordance with the change in the field current value If.

Therefore, the demagnetization determination section 36 corrects the demagnetization determination value Thφ based on the field current value If. For example, the reference field current value If0 when setting the q-axis voltage command value Vqoth for judgment value setting is set in advance for each operating point of the rotational angular velocity ω and the q-axis current value Iq (or torque command value). It is set. As shown in the following equation, the demagnetization determination unit 36 determines the amount of variation ΔIf (= If -If0) multiplied by the inductance Lf of the field winding, and the reference demagnetization is set using the q-axis voltage command value Vqoth for setting the judgment value according to equations (23) and (24), etc. The demagnetization determination value Thφ is corrected by adding it to the determination value Thφ. Here, the field current value If is the field current detection value Ifs or the field current command value Ifo.

<Other embodiments>
In the above embodiment, the rotor 14 was provided with the permanent magnets 12 and the field windings 7. However, the rotor 14 may not be provided with the field winding 7 but may be provided with the permanent magnets 12.

In each of the above embodiments, two sets of armature windings were provided. However, one or more sets of armature windings may be provided.

In each of the above embodiments, each set was provided with three-phase armature windings. However, each set may be provided with armature windings of multiple phases other than three phases (for example, two phases, four phases).

Although this application describes exemplary embodiments, the various features, aspects, and functions described in the embodiments are not limited to the application of particular embodiments, and may be used alone or It is applicable to the embodiments in various combinations. Accordingly, countless variations not illustrated are envisioned within the scope of the technology disclosed herein. For example, this includes cases in which at least one component is modified, added, or omitted.

1 AC rotating machine, 7 Field winding, 12 Permanent magnet, 30 AC rotating machine control device, 31 Rotation detection unit, 32 Current detection unit, 33 Voltage command value calculation unit, 34 Switching control unit, 35 Interlinkage magnetic flux estimation Section, 36 Demagnetization judgment section, 37 Winding temperature acquisition section, Iq q-axis current value, Ra resistance value, Ra_max maximum value of resistance value variation range, Ra_min minimum value of resistance value variation range, Rath judgment value setting resistance value, Thφ demagnetization judgment value, ThφH large side demagnetization judgment value, ThφL small side demagnetization judgment value, Vqo q-axis voltage command value, Vqo_max maximum value of the variation range of q-axis voltage command value , Vqo_min: Minimum value of the variation range of the voltage command value on the q-axis, Vqoth: Voltage command value on the q-axis for setting the judgment value, φd: Estimated value of armature linkage flux on the d-axis, φdH: Armature on the larger side of the d-axis Estimated value of magnetic flux linkage, φdL Estimated value of armature magnetic flux linkage on the smaller d-axis, φdth Armature magnetic flux linkage on the d-axis for setting the judgment value, ω Rotational angular velocity

Claims

A control device for an AC rotating machine that controls, via a power converter, an AC rotating machine that has a rotor provided with magnets and a stator provided with m sets of armature windings (m is a natural number of 1 or more). And,
a rotation detection unit that detects a rotational angular velocity in electrical angle of the rotor;
a voltage command value calculation unit that calculates a voltage command value for each group;
a switching control unit that applies voltage to the armature winding by turning on and off a switching element included in the power converter based on the voltage command value for each set;
Based on m sets of the voltage command values, m sets of current values of the armature windings, resistance values of the armature windings, and the rotational angular velocity, the interlinkage magnetic flux interlinking with the armature windings is determined. A flux linkage estimator to estimate;
a demagnetization determination unit that determines whether demagnetization of the magnet has occurred based on a comparison result between the estimated value of the flux linkage and the demagnetization determination value;
A control device for AC rotating machines equipped with
m is a natural number of 2 or more,
The flux linkage estimator uses an average value of the m sets of the voltage command values as the m sets of the voltage command values, and uses an average value of the m sets of the current values as the m sets of the current values. 2. A control device for an AC rotating machine according to item 1.
The rotor has a field winding,
The switching control unit turns on and off a switching element for the field winding included in the power converter to apply a voltage to the field winding,
The control device for an AC rotating machine according to claim 1 or 2, wherein the estimated value of the flux linkage estimated by the flux linkage estimator includes flux linkage generated by the field winding.
Further comprising a winding temperature acquisition unit that detects or estimates and acquires the temperature of the armature winding,
The flux linkage estimation unit estimates the resistance value based on the acquired value of the temperature of the armature winding, and estimates the flux linkage using the estimated value of the resistance value as the resistance value. The control device for an AC rotating machine according to any one of claims 1 to 3.
The flux linkage estimation unit uses a q-axis voltage command value as the voltage command value, uses a q-axis current value as the current value, and uses a q-axis current value as the estimated value of the armature winding as the estimated value of the flux linkage. Estimate the d-axis magnetic flux linkage to
The alternating current according to any one of claims 1 to 4, wherein the d-axis is set in the direction of the north pole of the magnet, and the q-axis is set in a direction 90 degrees electrical angle ahead of the d-axis. Control device for rotating machines.
The flux linkage estimator is configured to calculate the flux linkage of the d-axis by setting the voltage command value of the q-axis as Vqo, setting the current value of the q-axis as Iq, setting the resistance value as Ra, and setting the rotational angular velocity as ω. As φd,
φd=(Vqo-Ra×Iq)/ω
The control device for an AC rotating machine according to claim 5, wherein the d-axis magnetic flux linkage is estimated using a calculation formula.
The flux linkage estimating unit estimates the flux linkage of the d-axis on the larger side using the minimum value of a preset variation range of the resistance value as the resistance value, and Estimate the interlinkage magnetic flux of the d-axis on the smaller side using the maximum value of the set variation range of the resistance value,
The demagnetization determination unit determines that demagnetization has not occurred when the estimated value of the interlinkage magnetic flux of the smaller d-axis is larger than the demagnetization determination value, and the demagnetization determination unit determines that demagnetization has not occurred. The control device for an AC rotating machine according to claim 6, wherein if the estimated value of the flux linkage is smaller than the demagnetization determination value, it is determined that demagnetization has occurred.
The flux linkage estimating unit estimates the flux linkage of the d-axis on the larger side using the minimum value of a preset variation range of the resistance value as the resistance value, and Estimate the interlinkage magnetic flux of the d-axis on the smaller side using the maximum value of the set variation range of the resistance value,
The demagnetization determination unit sets a larger demagnetization determination value and a smaller demagnetization determination value that is smaller than the larger demagnetization determination value,
If the estimated value of the d-axis magnetic flux linkage on the smaller side is larger than the demagnetization judgment value on the larger side, it is determined that demagnetization has not occurred, and the estimated value of the magnetic flux linkage on the larger side is determined. 7. The control device for an AC rotating machine according to claim 6, wherein it is determined that demagnetization is occurring when the demagnetization determination value is smaller than the smaller demagnetization determination value.
The demagnetization determination section includes:
The smaller demagnetization judgment value corresponding to the maximum value of the preset variation range of the resistance value and the larger demagnetization judgment value that is larger than the smaller demagnetization judgment value are set. do, or
the larger side demagnetization judgment value corresponding to the minimum value of the preset resistance value variation range; and the smaller side demagnetization judgment value corresponding to the preset maximum value of the resistance value variation range. and set
The control device for an AC rotating machine according to claim 8, wherein when the estimated value of the smaller side flux linkage is smaller than the smaller side demagnetization determination value, it is determined that demagnetization has occurred.
The demagnetization determination section includes:
the larger demagnetization judgment value corresponding to the minimum value of the preset variation range of the resistance value; and the smaller demagnetization judgment value that is smaller than the larger demagnetization judgment value. set or
the larger side demagnetization judgment value corresponding to the minimum value of the preset resistance value variation range; and the smaller side demagnetization judgment value corresponding to the preset maximum value of the resistance value variation range. and set
The control device for an AC rotating machine according to claim 8 or 9, which determines that demagnetization has not occurred when the estimated value of the larger side magnetic flux linkage is larger than the larger side demagnetization determination value. .
The demagnetization determining unit determines the d value for determining a determination value based on the voltage command value of the q-axis for determining a determination value, the current value for the q-axis, the resistance value for determining a determination value, and the rotational angular velocity. 6. The control device for an AC rotating machine according to claim 5, wherein the demagnetization determination value is set based on the calculated magnetic flux linkage of the d-axis for setting the determination value by calculating the magnetic flux linkage of the shaft.
The demagnetization determination unit sets the voltage command value of the q-axis for setting a judgment value to Vqoth, the current value for the q-axis to Iq, the resistance value for setting a judgment value to Rath, and the rotational angular velocity to ω. and the interlinking magnetic flux of the d-axis for setting the judgment value interlinking with the armature winding is φdth,
φdth=(Vqoth-Rath×Iq)/ω
The d-axis magnetic flux linkage for setting the determination value is calculated using the calculation formula, and the demagnetization determination value is set based on the calculated d-axis magnetic flux linkage for setting the determination value. 5. The control device for an AC rotating machine according to 5.
The demagnetization determination unit calculates the d-axis magnetic flux linkage for setting the judgment value, using a minimum value in a preset variation range of the resistance value as the resistance value for setting the judgment value, setting the demagnetization judgment value based on the d-axis interlinkage magnetic flux for setting the judgment value;
If the estimated value of the magnetic flux linkage is larger than the demagnetization judgment value, it is determined that demagnetization has not occurred, and if the estimated value of the magnetic flux linkage is smaller than the demagnetization judgment value. The control device for an AC rotating machine according to claim 11 or 12, wherein the control device determines that demagnetization has occurred.
The demagnetization determination unit calculates a d-axis magnetic flux linkage for setting the judgment value, using a maximum value of a preset variation range of the resistance value as the resistance value for setting the judgment value, setting the demagnetization judgment value based on the d-axis interlinkage magnetic flux for setting the judgment value;
If the estimated value of the magnetic flux linkage is larger than the demagnetization judgment value, it is determined that demagnetization has not occurred, and if the estimated value of the magnetic flux linkage is smaller than the demagnetization judgment value. The control device for an AC rotating machine according to claim 11 or 12, wherein the control device determines that demagnetization has occurred.
Further comprising a winding temperature acquisition unit that detects or estimates and acquires the temperature of the armature winding,
The demagnetization determination unit estimates the resistance value based on the obtained value of the temperature of the armature winding, and uses the estimated value of the resistance value as the resistance value for setting the determination value. calculating a d-axis magnetic flux linkage for value setting, and setting the demagnetization judgment value based on the calculated d-axis magnetic flux linkage for setting the judgment value;
If the estimated value of the magnetic flux linkage is larger than the demagnetization judgment value, it is determined that demagnetization has not occurred, and if the estimated value of the magnetic flux linkage is smaller than the demagnetization judgment value. The control device for an AC rotating machine according to claim 11 or 12, wherein the control device determines that demagnetization has occurred.
The demagnetization determination section includes:
As the resistance value for setting the judgment value, the minimum value of the variation range of the resistance value set in advance is used to calculate the d-axis linkage flux for setting the judgment value on the larger side, and The d-axis interlinkage magnetic flux for judgment value setting is set as the larger demagnetization judgment value, and a value smaller than the larger demagnetization judgment value is set as the smaller demagnetization judgment value. , or
As the resistance value for setting the judgment value, the minimum value of the variation range of the resistance value set in advance is used to calculate the d-axis linkage flux for setting the judgment value on the larger side, and The d-axis interlinkage magnetic flux for setting the judgment value is set as the larger demagnetization judgment value, and the maximum value of the preset variation range of the resistance value is used as the resistance value for setting the judgment value. calculate the d-axis magnetic flux linkage for setting the judgment value on the small side, and set the d-axis magnetic flux linkage for setting the judgment value on the small side as the demagnetization judgment value on the small side; Or
As the resistance value for setting the judgment value, the maximum value of the variation range of the resistance value set in advance is used to calculate the d-axis interlinkage flux for setting the judgment value on the smaller side. A d-axis interlinkage magnetic flux for setting a determination value is set as the smaller demagnetization determination value, and a value larger than the smaller demagnetization determination value is set as the larger demagnetization determination value. ,
If the estimated value of the flux linkage is larger than the larger demagnetization judgment value, it is determined that no demagnetization has occurred, and the estimated value of the flux linkage is determined to be larger than the smaller demagnetization judgment value. The control device for an AC rotating machine according to claim 11 or 12, wherein if the value is smaller than the value, it is determined that demagnetization has occurred.
Further comprising a winding temperature acquisition unit that detects or estimates and acquires the temperature of the armature winding,
The demagnetization determination unit estimates the resistance value based on the obtained value of the temperature of the armature winding, and uses the estimated value of the resistance value as the resistance value for setting the determination value. Calculate the d-axis magnetic flux linkage for value setting, and based on the calculated d-axis magnetic flux linkage for judgment value setting, the larger demagnetization judgment value and the larger demagnetization judgment value. Set the demagnetization judgment value on the smaller side of the smaller value,
If the estimated value of the flux linkage is larger than the larger demagnetization judgment value, it is determined that no demagnetization has occurred, and the estimated value of the flux linkage is determined to be larger than the smaller demagnetization judgment value. The control device for an AC rotating machine according to claim 11 or 12, wherein if the value is smaller than the value, it is determined that demagnetization has occurred.
The demagnetization determination unit uses a preset voltage command value of the q-axis when demagnetization does not occur as the voltage command value of the q-axis for setting the determination value, and determines the voltage command value of the d-axis for setting the determination value. The control device for an AC rotating machine according to any one of claims 11 to 17, wherein the control device calculates a magnetic flux linkage.
The demagnetization determination section uses a preset maximum value of a variation range of the q-axis voltage command value when demagnetization does not occur as the q-axis voltage command value for setting the determination value, and performs the determination. The control device for an AC rotating machine according to claim 13, wherein the control device for an AC rotating machine calculates a d-axis interlinkage magnetic flux for value setting.
The demagnetization determination section uses a preset minimum value of a variation range of the q-axis voltage command value when demagnetization does not occur as the q-axis voltage command value for setting the determination value, and performs the determination. The control device for an AC rotating machine according to claim 14, wherein the control device for an AC rotating machine calculates a d-axis interlinkage magnetic flux for value setting.
The control device for an AC rotating machine according to any one of claims 1 to 20, wherein the demagnetization determination unit corrects the demagnetization determination value based on the rotational angular velocity.
The demagnetization determination unit determines whether or not to perform a demagnetization determination based on the rotational angular velocity, the voltage command value, and the current value, and when it is determined that the demagnetization determination is to be performed, the The control device for an AC rotating machine according to any one of claims 1 to 21, which determines whether demagnetization of the magnet has occurred.
The rotor has a field winding,
The switching control unit turns on and off a switching element for the field winding included in the power converter to apply a voltage to the field winding,
The AC rotating machine according to any one of claims 1 to 22, wherein the demagnetization determination unit corrects the demagnetization determination value based on a field current value that is a current value flowing through the field winding. control device.