WO2022244204A1 - 回転機制御装置、機械学習装置および推論装置 - Google Patents
回転機制御装置、機械学習装置および推論装置 Download PDFInfo
- Publication number
- WO2022244204A1 WO2022244204A1 PCT/JP2021/019223 JP2021019223W WO2022244204A1 WO 2022244204 A1 WO2022244204 A1 WO 2022244204A1 JP 2021019223 W JP2021019223 W JP 2021019223W WO 2022244204 A1 WO2022244204 A1 WO 2022244204A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- value
- current
- magnetic flux
- control period
- length
- Prior art date
Links
- 238000010801 machine learning Methods 0.000 title claims description 62
- 230000004907 flux Effects 0.000 claims abstract description 255
- 238000001514 detection method Methods 0.000 claims abstract description 84
- 238000004364 calculation method Methods 0.000 claims description 78
- 230000010349 pulsation Effects 0.000 claims description 67
- 230000006870 function Effects 0.000 claims description 30
- 230000009467 reduction Effects 0.000 claims description 18
- 230000001360 synchronised effect Effects 0.000 description 58
- 230000009471 action Effects 0.000 description 35
- 238000000034 method Methods 0.000 description 28
- 238000010586 diagram Methods 0.000 description 19
- 238000012545 processing Methods 0.000 description 16
- 230000005284 excitation Effects 0.000 description 12
- 230000002787 reinforcement Effects 0.000 description 8
- 230000004048 modification Effects 0.000 description 7
- 238000012986 modification Methods 0.000 description 7
- 230000004913 activation Effects 0.000 description 6
- 230000007423 decrease Effects 0.000 description 5
- 230000007704 transition Effects 0.000 description 5
- 238000001914 filtration Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 3
- 230000003542 behavioural effect Effects 0.000 description 2
- 239000003795 chemical substances by application Substances 0.000 description 2
- 230000001965 increasing effect Effects 0.000 description 2
- 230000000087 stabilizing effect Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000001939 inductive effect Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000001172 regenerating effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
- H02P21/22—Current control, e.g. using a current control loop
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P1/00—Arrangements for starting electric motors or dynamo-electric converters
- H02P1/02—Details of starting control
- H02P1/029—Restarting, e.g. after power failure
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
- H02P21/0003—Control strategies in general, e.g. linear type, e.g. P, PI, PID, using robust control
- H02P21/0014—Control strategies in general, e.g. linear type, e.g. P, PI, PID, using robust control using neural networks
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
- H02P21/0003—Control strategies in general, e.g. linear type, e.g. P, PI, PID, using robust control
- H02P21/0025—Control strategies in general, e.g. linear type, e.g. P, PI, PID, using robust control implementing a off line learning phase to determine and store useful data for on-line control
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
- H02P21/14—Estimation or adaptation of machine parameters, e.g. flux, current or voltage
- H02P21/141—Flux estimation
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
- H02P21/14—Estimation or adaptation of machine parameters, e.g. flux, current or voltage
- H02P21/18—Estimation of position or speed
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
- H02P21/34—Arrangements for starting
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P25/00—Arrangements or methods for the control of AC motors characterised by the kind of AC motor or by structural details
- H02P25/02—Arrangements or methods for the control of AC motors characterised by the kind of AC motor or by structural details characterised by the kind of motor
- H02P25/022—Synchronous motors
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P2207/00—Indexing scheme relating to controlling arrangements characterised by the type of motor
- H02P2207/05—Synchronous machines, e.g. with permanent magnets or DC excitation
Definitions
- the present disclosure relates to a rotating machine control device that controls a rotating machine, a machine learning device, and an inference device.
- a position sensor was used to detect the rotor position in an AC rotating machine.
- the phase of the magnetic flux vector is obtained by detecting the induced voltage generated by the rotation of the magnetic flux vector in the AC rotating machine, and the rotor is synchronized with the phase of the magnetic flux vector.
- Sensorless vector control for position determination is known. According to the sensorless vector control, the AC rotating machine does not require a position sensor, which is a precision device, so failures can be reduced and costs can be reduced.
- the rotating machine control device In sensorless vector control, the rotating machine control device generally estimates the induced voltage based on the voltage command value input to the power converter or the detected value of the current flowing through the AC rotating machine. When estimating the induced voltage based on the voltage command value or current detection value, the rotating machine control device cannot detect the phase and frequency of the magnetic flux vector when the AC rotating machine is in a free-running state.
- the free run state is a state in which the rotor continues to rotate due to inertia when the power supply from the power converter is cut off while the rotor is rotating.
- the phase and frequency cannot be detected, and appropriate current control becomes difficult. Due to the occurrence of pulsation at , it becomes difficult to start from the free-running state. If the generated torque is regenerative torque, a sudden rise in the DC bus voltage of the power converter may cause problems in the power converter.
- Patent Literature 1 discloses a rotating machine control device that determines a voltage command value by performing current control to set the current command value to zero during a startup control period.
- the startup control period it is desirable to set a voltage command value that can reduce the generation of excessive torque, the generation of pulsation, and the rise of the DC bus voltage as much as possible. Therefore, it is desirable to determine the voltage command value by current control using a very small current command value.
- the current command value during the startup control period is small, the generation of excessive torque can be reduced, while the magnetic flux generated by the excitation in the AC rotating machine also becomes small. The smaller the magnetic flux, the more unstable the magnetic flux, which requires a certain time constant until the estimated phase or frequency stabilizes, thus lengthening the start-up control period. Also, the small and unstable magnetic flux is likely to cause unintended torque and current or frequency pulsation.
- the present disclosure has been made in view of the above, and can shorten the time required to start a rotating machine from a free-running state, and can reduce the occurrence of unintended torque and current or frequency pulsation. It is an object of the present invention to obtain a rotating machine control device that
- a rotating machine control device includes a current detector that detects an alternating current flowing through a rotating machine and outputs a current detection value, and a voltage command value based on a voltage command value.
- a power converter that supplies power to the rotating machine by applying an alternating voltage
- a current controller that adjusts the voltage command value so that the detected current value matches the current command value, and an estimation of the amplitude of the magnetic flux vector in the rotating machine.
- An estimator that obtains a magnetic flux estimated value that is a value, and a starting control period from when the power supply from the power converter is cut off and the rotating machine starts to rotate by inertia to when the power supply is restarted is set. and a magnetic flux controller that adjusts the current command value so that the magnetic flux estimated value matches the magnetic flux command value.
- the rotating machine control device can shorten the time required to start the rotating machine from the free-running state, and can reduce the occurrence of unintended torque and current or frequency pulsation.
- FIG. 1 is a diagram showing a configuration example of a rotating machine control device according to a first embodiment
- FIG. FIG. 2 is a diagram showing a configuration example of a controller included in the rotating machine control device according to the first embodiment
- FIG. 3 is a diagram showing a configuration example of an estimator included in the rotating machine control device according to the first embodiment
- FIG. 4 is a diagram for explaining the operation of the rotating machine control device according to the first embodiment during a startup control period
- FIG. 7 is a diagram showing a configuration example of an estimator included in the rotating machine control device according to the second embodiment
- FIG. 4 is a diagram showing a configuration example of hardware that realizes the voltage commander of the rotating machine control device according to the first to third embodiments;
- FIG. 11 is a diagram showing a machine learning device and a rotating machine control device according to a fourth embodiment; 10 is a flowchart showing the processing procedure of the machine learning device according to the fourth embodiment;
- FIG. 11 shows an inference device and a rotating machine control device according to a fifth embodiment;
- 14 is a flow chart showing a processing procedure of an inference device according to a fifth embodiment;
- FIG. 12 shows an inference device and a rotating machine control device according to a modification of the fifth embodiment;
- FIG. 1 is a diagram showing a configuration example of a rotating machine control device 20 according to a first embodiment.
- the rotating machine control device 20 controls a synchronous machine 21 that is an AC rotating machine.
- the two axes in the fixed coordinate system are called the ⁇ -axis and ⁇ -axis
- the two axes in the rotating coordinate system are called the d-axis and the q-axis.
- Embodiment 1 describes a case where the control phase is synchronized with the phase of the primary magnetic flux vector.
- the d-axis and the q-axis in the case of coordinate conversion to the rotating coordinates with the direction of the primary magnetic flux vector as the reference d-axis are referred to as the ds-axis and the qs-axis, respectively.
- the ds-axis is the axis in the direction of the primary flux vector.
- the qs axis is the axis perpendicular to the primary magnetic flux vector.
- the primary magnetic flux represents the stator magnetic flux.
- a secondary magnetic flux represents a magnetic flux corresponding to the magnetic field of the synchronous machine 21 .
- the secondary magnetic flux is the magnetic flux generated by the magnetic field of the rotor.
- the secondary magnetic flux is the combined magnetic flux generated by the magnetic field of the rotor and the magnetic flux produced by the saliency of the rotor. If the synchronous machine 21 is a synchronous machine that does not have a magnetic field in the rotor, such as a reluctance synchronous machine, the secondary magnetic flux is the magnetic flux produced by the saliency of the rotor.
- the rotating machine control device 20 includes a voltage commander 10 that outputs voltage command values vus * , vvs * , and vws * based on current detection values ius, ivs , and iws , and a voltage command value vus * , v vs * and v ws * , the power converter 11 supplies power to the synchronous machine 21 by applying an AC voltage based on , v vs * and v ws *, and the AC current flowing through the synchronous machine 21 is detected to detect current values i us , i vs , and a current detector 12 which outputs i_ws .
- the voltage commander 10 has a controller 1 that generates voltage command values vus * , vvs * , vws * , and an estimator 2 that estimates magnetic flux, phase and frequency.
- the estimator 2 obtains a magnetic flux estimated value
- the estimator 2 obtains an estimated phase value ⁇ , which is an estimated value of the phase of the magnetic flux vector.
- the estimator 2 obtains a frequency estimate ⁇ , which is an estimate of the frequency of the magnetic flux vector.
- and the phase estimated value ⁇ output from the estimator 2 are input to the controller 1 and used to control the synchronous machine 21 during the startup control period.
- the phase estimated value ⁇ and frequency estimated value ⁇ output from the estimator 2 can also be used for control when the synchronous machine 21 returns from the free-running state to the normal driving state.
- Variables marked with ' ⁇ ' represent estimated values.
- FIG. 2 is a diagram showing a configuration example of the controller 1 included in the rotating machine control device 20 according to the first embodiment.
- the controller 1 has a three-phase dq converter 31 , a dq three-phase converter 32 , a current controller 33 , a switch 34 and a magnetic flux controller 35 .
- the three-phase dq converter 31 receives the current detection values i us , i vs , and i ws and the phase estimation value ⁇ .
- the three-phase dq converter 31 outputs a ds -axis current detection value ids and a qs -axis current detection value iqs by coordinate transformation of the current detection values i us , i vs , and i ws based on the phase estimation value ⁇ . .
- the current controller 33 receives the ds-axis current detection value ids, the qs-axis current detection value iqs , the ds -axis current command value ids * , and the qs-axis current command value iqs * .
- the current controller 33 controls the ds-axis current detection value ids such that the ds-axis current detection value ids matches the ds -axis current command value ids * and the qs-axis current detection value iqs matches the qs -axis current command value iqs * .
- a voltage command value v ds * and a qs-axis voltage command value v qs * are obtained.
- the current controller 33 outputs a ds-axis voltage command value v ds * and a qs-axis voltage command value v qs * . That is, the current controller 33 adjusts the voltage command value so that the detected current value matches the current command value.
- the ds-axis voltage command value v ds * , the qs-axis voltage command value v qs * , and the phase estimation value ⁇ are input to the dq three-phase converter 32 .
- the dq three-phase converter 32 converts the ds-axis voltage command value v ds * and the qs-axis voltage command value v qs * based on the phase estimation value ⁇ into the voltage command values v us * , v vs * , v Output ws * .
- the voltage commander 10 outputs voltage command values v us * , v vs * , v ws * to the power converter 11 .
- and a preset magnetic flux command value ⁇ * are input to the magnetic flux controller 35 .
- the magnetic flux controller 35 obtains an exciting current command value such that the magnetic flux estimated value
- the magnetic flux controller 35 outputs a d-axis excitation current command value.
- the switch 34 converts one of the d -axis current command value id * , which is a preset excitation current command value, and the d-axis excitation current command value output from the magnetic flux controller 35, to the ds-axis current command value. Output as i ds * .
- Information indicating the current control period T1 and information indicating the magnetic flux control period T2 are input to the switch 34 .
- the switch 34 selects between the d-axis current command value i d * and the d-axis excitation current command value output from the magnetic flux controller 35 based on the information on the current control period T1 and the information on the magnetic flux control period T2. Choose one.
- the current control period T1 and the magnetic flux control period T2 will be described later.
- the ds-axis current command value i ds * is input to the current controller 33 .
- FIG. 3 is a diagram showing a configuration example of the estimator 2 included in the rotating machine control device 20 according to the first embodiment.
- the estimator 2 operates during the entire startup control period.
- the estimator 2 receives voltage command values vus * , vvs * , vws * and current detection values ius, ivs , iws .
- One of the two three-phase dq converters in the estimator 2 converts the voltage command values vus * , vvs * , vws * based on the phase estimation value ⁇ into the ds-axis voltage
- a command value v ds * and a qs-axis voltage command value v qs * are output.
- the other of the two three-phase dq converters in the estimator 2 converts the current detection values i us , i vs , and i ws based on the phase estimation value ⁇ into the ds-axis current detection value i It outputs the ds and qs axis current detection values i qs .
- a ds -axis primary voltage vds and a qs -axis primary voltage vqs of the synchronous machine 21 are represented by the following equation (1).
- ⁇ ds is the ds -axis primary magnetic flux
- ⁇ qs is the qs-axis primary magnetic flux
- ids is the ds -axis primary current
- iqs is the qs-axis primary current
- Rs is the primary phase resistance
- ⁇ s is the frequency of the primary magnetic flux vector
- s is the Laplacian operator.
- of the primary magnetic flux vector is equal to the ds-axis primary magnetic flux ⁇ ds .
- of the primary magnetic flux vector is expressed by the following equation (2).
- the frequency ⁇ s of the primary magnetic flux vector is expressed by the following equation (3).
- the estimator 2 having the configuration in FIG. 3, performs the calculation shown in Equation (3) from the voltage command value and the current detection value to estimate the frequency ⁇ s of the primary magnetic flux vector.
- the phase ⁇ s of the primary magnetic flux vector is represented by the following equation (4).
- the estimator 2 having the configuration of FIG. 3, performs the calculation shown in Equation (4) from the voltage command value and the current detection value to estimate the phase ⁇ s of the primary magnetic flux vector.
- the estimator 2 outputs a flux estimate
- the estimator 2 outputs a frequency estimate ⁇ which is an estimate of the frequency of the primary flux vector ⁇ s .
- the estimator 2 outputs a phase estimate ⁇ ⁇ which is an estimate ⁇ s ⁇ of the phase of the primary magnetic flux vector.
- the term of the voltage command value may include a term for correcting the voltage error of the power converter 11 .
- FIG. 4 is a diagram for explaining the operation of the rotating machine control device 20 according to the first embodiment during the startup control period.
- the horizontal axis represents time.
- FIG. 4 shows the state transition of the rotating machine control device 20 and the details of the startup control period.
- a driving state refers to a state in which the synchronous machine 21 is driven under normal control by the rotating machine control device 20 .
- the state of the synchronous machine 21 changes from the driving state to the free-running state, and then returns to the driving state.
- the startup control period consists of a current control period T1 in which current control is performed based on the d -axis current command value id * , which is a predetermined command value, and a current control period T1 based on the d-axis excitation current command value output by the magnetic flux controller 35. and a flux control period T2 during which control is performed.
- the current control period T1 is a period from t0 to t1.
- the magnetic flux control period T2 is a period from t1 to t2, which is the end of the activation control period.
- the synchronous machine 21 transitions from the free-running state to the driving state.
- the switch 34 selects the d-axis current command value i d * according to information on the current control period T1.
- the switch 34 outputs a ds-axis current command value i ds * that is the d-axis current command value i d * .
- the d-axis current command value i d * is a value that indicates a minute excitation current.
- the rotating machine control device 20 causes the synchronous machine 21 to generate magnetic flux by causing a minute excitation current to flow through the synchronous machine 21 during the current control period T1.
- the current control period T1 is a preliminary operation period for smoothly starting the magnetic flux control in the magnetic flux control period T2 following the current control period T1. Therefore, basically, in the startup control period, the magnetic flux control period T2 is set to be longer than the current control period T1.
- the switch 34 selects the d-axis excitation current command value output by the magnetic flux controller 35 according to the information on the magnetic flux control period T2.
- the switch 34 outputs a ds -axis current command value ids * , which is the d-axis excitation current command value output by the magnetic flux controller 35 .
- the phase estimate ⁇ obtained by the estimator 2 during the start-up control period is used as the initial phase value in normal control from t2.
- the frequency estimate ⁇ obtained by the estimator 2 during the start-up control period is used as the initial value of the frequency in normal control from t2.
- the phase estimated value ⁇ and frequency estimated value ⁇ are the phase and frequency of the primary magnetic flux vector. Since the synchronous machine 21 has no load in the free-running state, the direction of the primary magnetic flux vector is the same as the direction of the secondary magnetic flux vector in the free-running state. Therefore, even if a control method based on the secondary magnetic flux vector is adopted for control in the drive state, the phase It does not matter if the estimate ⁇ and the frequency estimate ⁇ are used.
- the entire activation control period is the current control period.
- the current command value in order to minimize generation of excessive torque, generation of pulsation, and rise in the DC bus voltage, the current command value must be set to a value that indicates a minute excitation current. In this case, a certain time constant is required until the estimated phase or frequency stabilizes, so the activation control period becomes longer.
- the phase and frequency obtained from the magnetic flux vector of the synchronous machine 21 or the induced voltage generated by the rotation of the magnetic flux vector are likely to be unstable due to the influence of disturbances and the like. Phase and frequency instability can easily lead to unintended torque generation and current or frequency pulsation. Also, the DC bus voltage may rise.
- the rotating machine control device 20 provides the current control period T1 before the magnetic flux control period T2 in the startup control period, so that the magnetic flux estimated value
- the rotating machine control device 20 performs magnetic flux control using the magnetic flux estimated value
- the rotating machine control device 20 switches from the current control period T1 to the magnetic flux control period T2 at an appropriate timing, and performs magnetic flux control using the magnetic flux estimated value
- the rotating machine control device 20 uses the d-axis excitation current command value output by the magnetic flux controller 35 in the magnetic flux control period T2, thereby increasing the magnetic flux vector at an early stage. value and stabilize the magnetic flux vector at an early stage. Therefore, the rotating machine control device 20 can shorten the time until the estimated phase and frequency are stabilized. As a result, the synchronous machine 21 can transition from the free-running state to the driving state in a short startup control period. In addition, the synchronous machine 21 can reduce phase and frequency fluctuations due to disturbances and the like. By stabilizing phase and frequency early, unintended torque generation and current or frequency pulsation can be reduced. Further, the rotating machine control device 20 can reduce the increase in the DC bus voltage.
- the rotating machine control device 20 can shorten the time required to start the rotating machine from the free-running state, and can reduce the occurrence of unintended torque and current or frequency pulsation. .
- Embodiment 2 In the first embodiment, an example of obtaining the magnetic flux estimation value
- FIG. 5 is a diagram showing a configuration example of the estimator 2A included in the rotating machine control device 20 according to the second embodiment.
- the configuration of the rotating machine control device 20 according to the second embodiment other than the estimator 2A is the same as that of the rotating machine control device 20 according to the first embodiment.
- descriptions overlapping those of the first embodiment are omitted.
- the rotating machine control device 20 synchronizes the control phase with the phase of the primary magnetic flux vector.
- the ⁇ -axis and ⁇ -axis in the case of performing coordinate conversion to fixed coordinates with the direction of the primary magnetic flux vector as the reference ⁇ -axis are referred to as the ⁇ s-axis and the ⁇ s-axis, respectively.
- the ⁇ s axis is the axis in the direction of the primary magnetic flux vector.
- the ⁇ s axis is the axis perpendicular to the primary magnetic flux vector.
- the voltage command values vus * , vvs * , vws * and the current detection values ius, ivs, iws are input to the estimator 2A.
- One of the two three-phase ⁇ converters in the estimator 2A converts the voltage command values vus * , vvs * , and vws * into the ⁇ s-axis voltage command values v ⁇ s * and the ⁇ s-axis Output the voltage command value v ⁇ s * .
- the other of the two three-phase ⁇ converters in the estimator 2A converts the current detection values i us , i vs , and i ws into the ⁇ s-axis current detection value i ⁇ s and the ⁇ s-axis current detection value i Output ⁇ s .
- the following formula (5) is a formula expressing the above formula (1) with the ⁇ -axis and the ⁇ -axis.
- v ⁇ s is the ⁇ s-axis primary voltage
- v ⁇ s is the ⁇ s-axis primary voltage
- i ⁇ s is the ⁇ s-axis primary current
- i ⁇ s is the ⁇ s-axis primary current
- ⁇ ⁇ s is the ⁇ s-axis primary magnetic flux
- ⁇ ⁇ s is the ⁇ s-axis primary magnetic flux.
- Equation (6) is the integration of the AC quantity, unlike Equation (2) above. Therefore, in order to avoid accumulating a DC component error such as a detection offset, the calculation may be combined with a high-pass filter having a cutoff frequency ⁇ k .
- the cutoff frequency ⁇ k is a value less than the frequency estimate ⁇ and is determined by multiplying the frequency estimate ⁇ by a constant.
- a constant is, for example, a value of 1/3 or less than 1/3.
- FIG. 5 shows a configuration example of the estimator 2A when a high-pass filter is combined.
- of the primary magnetic flux vector is expressed by the following equation (8).
- the estimator 2A having the configuration in FIG. 5, performs the calculations shown in equations (7) and (8) from the voltage command value and current detection value to estimate the amplitude
- the estimator 2A may estimate the amplitude
- the phase ⁇ s of the primary magnetic flux vector is represented by the following equation (9).
- the estimator 2A having the configuration of FIG. 5, performs the calculations shown in equations (7) and (9) from the voltage command value and current detection value to estimate the phase ⁇ s of the primary magnetic flux vector.
- the estimator 2A may estimate the phase ⁇ s of the primary magnetic flux vector by performing the calculations shown in Equations (6) and (9).
- the frequency ⁇ s of the primary magnetic flux vector is expressed by the following equation (10).
- the estimator 2A having the configuration shown in FIG. 5, performs the calculations shown in equations (7) and (10) from the voltage command value and current detection value to estimate the frequency ⁇ s of the primary magnetic flux vector.
- the estimator 2A may estimate the frequency ⁇ s of the primary magnetic flux vector by performing the calculations shown in Equations (6) and (10).
- the estimator 2A outputs the magnetic flux estimate
- the estimator 2A outputs a frequency estimate ⁇ which is an estimate of the frequency of the primary magnetic flux vector ⁇ ⁇ .
- the estimator 2A outputs a phase estimate ⁇ ⁇ , which is an estimate ⁇ s ⁇ of the phase of the primary magnetic flux vector.
- the term for the voltage command value may include a term for correcting the voltage error of the power converter 11 .
- the rotating machine control device 20 can shorten the time required to start the rotating machine from the free-run state, generate unintended torque, Alternatively, it is possible to reduce the occurrence of frequency pulsation.
- Embodiment 3 In Embodiments 1 and 2, the example in which the magnetic flux estimated value
- FIG. 6 is a diagram showing a configuration example of the estimator 2B included in the rotating machine control device 20 according to the third embodiment.
- the configuration of the rotating machine control device 20 according to the third embodiment other than the estimator 2B is the same as that of the rotating machine control device 20 according to the first embodiment.
- descriptions overlapping those of the first embodiment are omitted.
- the rotating machine control device 20 synchronizes the control phase with the phase of the secondary magnetic flux vector.
- the ⁇ -axis and ⁇ -axis when the coordinate transformation is performed with the direction of the secondary magnetic flux vector as the reference ⁇ -axis are referred to as the ⁇ r-axis and the ⁇ r-axis, respectively.
- the ⁇ r axis is the axis in the direction of the secondary magnetic flux vector.
- the ⁇ r axis is the axis perpendicular to the secondary magnetic flux vector.
- the voltage command values vus * , vvs * , vws * and the current detection values ius, ivs, iws are input to the estimator 2B.
- One of the two three-phase ⁇ converters in the estimator 2B converts the voltage command values v us * , v vs * , and v ws * into the ⁇ s-axis voltage command values v ⁇ s * and the ⁇ s-axis Output the voltage command value v ⁇ s * .
- the other of the two three-phase ⁇ converters in the estimator 2B converts the current detection values i us , i vs , and i ws into the ⁇ s-axis current detection value i ⁇ s and the ⁇ s-axis current detection value i Output ⁇ s .
- FIG. 6 shows a configuration example of the estimator 2B when a high-pass filter is combined as in the second embodiment.
- a method for extracting the components synchronous with the ⁇ r-axis and the ⁇ r-axis from the primary magnetic flux vector shown in the above equation (7) a method according to a known technique can be used.
- Equation (11) shows an example of a method for extracting components synchronized with the ⁇ r-axis and the ⁇ r-axis.
- the ⁇ r-axis primary magnetic flux ⁇ ⁇ r and the ⁇ r-axis primary magnetic flux ⁇ ⁇ r are obtained by the equation (11).
- L qr is the inductance of the qr-axis component.
- the d-axis and the q-axis in the case of performing the coordinate conversion to the rotating coordinates with the direction of the secondary magnetic flux vector as the reference d-axis are referred to as the dr-axis and the qr-axis, respectively.
- of the secondary magnetic flux vector is expressed by the following equation (12).
- the estimator 2B estimates the amplitude
- the phase ⁇ r of the secondary magnetic flux vector is represented by the following equation (13).
- the estimator 2B estimates the phase ⁇ r of the secondary magnetic flux vector with the configuration of FIG.
- the frequency ⁇ r of the secondary magnetic flux vector is expressed by the following equation (14).
- the estimator 2B estimates the frequency ⁇ r of the secondary magnetic flux vector with the configuration of FIG.
- the estimator 2B outputs a magnetic flux estimate
- the estimator 2B outputs a frequency estimate ⁇ which is an estimate of the frequency of the secondary flux vector ⁇ r ⁇ .
- the estimator 2B outputs a phase estimate ⁇ ⁇ which is an estimate ⁇ r ⁇ of the phase of the secondary magnetic flux vector.
- the term for the voltage command value may include a term for correcting the voltage error of the power converter 11 .
- the phase estimated value ⁇ and frequency estimated value ⁇ are the phase and frequency of the secondary magnetic flux vector. Since the synchronous machine 21 has no load in the free-running state, the direction of the secondary magnetic flux vector is the same as the direction of the primary magnetic flux vector in the free-running state. Therefore, even if a control method based on the primary magnetic flux vector is adopted for control in the driving state, the phase It does not matter if the estimate ⁇ and the frequency estimate ⁇ are used.
- the rotating machine control device 20 can shorten the time until the estimated phase and frequency stabilize, as in the case of the first embodiment. Therefore, the synchronous machine 21 can transition from the free-running state to the driving state in a short startup control period. In addition, the synchronous machine 21 can reduce phase and frequency fluctuations due to disturbances and the like. By stabilizing phase and frequency early, unintended torque generation and current or frequency pulsation can be reduced. Further, the rotating machine control device 20 can reduce the increase in the DC bus voltage.
- the rotating machine control device 20 can shorten the time required to start the rotating machine from the free-running state, and can reduce the occurrence of unintended torque and current or frequency pulsation. .
- FIG. 7 is a diagram showing a configuration example of hardware that implements the voltage commander 10 of the rotating machine control device 20 according to the first to third embodiments.
- FIG. 7 shows a configuration example in which controller 1 and estimators 2, 2A and 2B of voltage commander 10 are realized by processing circuit 61 having processor 63 and memory 64.
- processing circuit 61 having processor 63 and memory 64.
- the processor 63 is a CPU (Central Processing Unit).
- the processor 63 may be an arithmetic unit, microprocessor, microcomputer, or DSP (Digital Signal Processor).
- the memory 64 is, for example, a volatile or It is a non-volatile semiconductor memory.
- the memory 64 stores programs for operating as the controller 1 and the estimators 2, 2A, and 2B.
- the controller 1 and the estimators 2, 2A and 2B can be realized by reading and executing this program by the processor 63.
- the programs stored in the memory 64 for operating as the controller 1 and the estimators 2, 2A, and 2B are stored in storage media such as CD (Compact Disc)-ROM, DVD (Digital Versatile Disc)-ROM, etc. It may be provided to a user or the like in a state in which it is written in the file, or may be provided via a network.
- the processor 63 outputs data such as calculation results to the volatile memory of the memory 64 .
- the processor 63 stores the data such as the calculation result by outputting the data to the auxiliary storage device via the volatile memory of the memory 64 .
- the input unit 62 is a circuit that receives an input signal to the voltage commander 10 from the outside.
- the input unit 62 receives the current detection values i us , i vs , i ws , information on the current control period T1, and information on the magnetic flux control period T2.
- the output unit 65 is a circuit that outputs the signal generated by the voltage commander 10 to the outside.
- the output unit 65 outputs voltage command values vus * , vvs * , and vws * .
- FIG. 7 is an example of hardware when the controller 1 and estimators 2, 2A, and 2B are implemented by a general-purpose processor 63 and memory 64.
- a dedicated processing circuit is used for control. 1 and estimators 2, 2A, 2B may be implemented. That is, the controller 1 and the estimators 2, 2A, and 2B may be realized by dedicated processing circuits.
- the dedicated processing circuit is a single circuit, a composite circuit, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a circuit combining these.
- a part of the controller 1 and the estimators 2, 2A and 2B may be realized by the processor 63 and the memory 64, and the rest may be realized by a dedicated processing circuit.
- Embodiment 4 describes a machine learning device that generates a learned model for inferring the length of the current control period T1 and the length of the magnetic flux control period T2.
- FIG. 8 is a diagram showing the machine learning device 40 and the rotary machine control device 20A according to the fourth embodiment.
- the same components as those in the first to third embodiments are denoted by the same reference numerals, and different configurations from those in the first to third embodiments will be mainly described.
- a machine learning device 40 shown in FIG. 8 is provided outside the rotating machine control device 20A.
- a rotating machine control device 20A is obtained by adding a voltage detector 15 to the rotating machine control device 20 according to any one of the first to third embodiments.
- the voltage detector 15 detects the DC bus voltage of the power converter 11 and outputs a bus voltage detection value Vdc .
- the machine learning device 40 reduces the rise of the DC bus voltage in the power converter 11, reduces the pulsation width of the DC bus voltage, and reduces the pulsation width of the frequency of the magnetic flux vector in the synchronous machine 21 for the rotating machine control device 20A.
- the length of the current control period T1 and the length of the magnetic flux control period T2 that enable reduction and reduction of the pulsation width of the current effective value of the current flowing through the synchronous machine 21 are learned.
- the machine learning device 40 has a learning data acquisition unit 41 , a reward calculation data acquisition unit 42 , a model generation unit 43 , and a learned model storage unit 46 .
- the machine learning device 40 performs at least the reduction of the rise in the DC bus voltage, the reduction of the pulsation width of the DC bus voltage, the reduction of the pulsation width of the frequency of the magnetic flux vector, and the reduction of the pulsation width of the current effective value. It is sufficient if the length of the current control period T1 and the length of the magnetic flux control period T2 that enable one are learned.
- the learning data acquisition unit 41 obtains the value indicating the length of the current control period T1, the value indicating the length of the magnetic flux control period T2, the estimated frequency value ⁇ , and the effective current value I rms from the voltage commander 10. Get from Hereinafter, the value indicating the length of the current control period T1 will be referred to as the value of the current control period T1. A value indicating the length of the magnetic flux control period T2 is referred to as a value of the magnetic flux control period T2. The learning data acquisition unit 41 acquires learning data including values of the current control period T1 and the magnetic flux control period T2, the estimated frequency value ⁇ , and the current effective value I rms .
- the current effective value Irms is a current effective value obtained from the current detection values ius , ivs , and iws , and is a current effective value based on at least one of the current detection values ius, ivs , and iws . .
- a three-phase dq converter 31 in the power converter 11 calculates a current rms value I rms from at least one of the current detection values i us , ivs and i ws .
- the learning data acquisition unit 41 acquires the current effective value I rms calculated by the three-phase dq converter 31 .
- the three-phase dq converter 31 may calculate the current effective value I rms from the ds -axis primary current ids or the qs-axis primary current i qs .
- the current effective value I rms is calculated by the three-phase dq converter 31, it is not limited to this.
- the current effective value I rms may be calculated by either an internal element of the rotating machine control device 20A or an external element of the rotating machine control device 20A. Elements internal to the machine learning device 40 may calculate the current rms value I rms .
- the learning data acquisition unit 41 acquires the current detection values ius, ivs, and iws , the ds -axis primary current ids, or the qs -axis primary current iqs, and calculates the current effective value Irms . Also good.
- the learning data acquisition unit 41 operates and acquires learning data when the synchronous machine 21 is in the driving state.
- the learning data acquisition unit 41 creates a data set in which learning data, which are each value of the current control period T1 and the magnetic flux control period T2, the estimated frequency value ⁇ , and the current effective value I rms are put together.
- the learning data acquisition unit 41 sends the created data set to the model generation unit 43 .
- the reward calculation data acquisition unit 42 acquires the estimated frequency value ⁇ and the effective current value I rms from the voltage commander 10 .
- the remuneration calculation data acquisition unit 42 acquires the bus voltage detection value Vdc from the voltage detector 15 .
- the remuneration calculation data acquisition unit 42 operates from when the synchronous machine 21 is in the driving state to when the synchronous machine 21 is in the free-running state.
- the remuneration calculation data acquisition unit 42 acquires the estimated frequency value ⁇ , the current effective value I rms , and the bus voltage detection value V dc .
- the remuneration calculation data acquisition unit 42 obtains the pulsation width V dc _rip of the bus voltage detection value V dc from the acquired bus voltage detection value V dc .
- the pulsation width V dc _rip is the amplitude of the waveform representing the time series change of the bus voltage detection value V dc .
- the remuneration calculation data acquisition unit 42 obtains the pulsation width ⁇ _rip of the frequency estimate ⁇ from the acquired frequency estimate ⁇ .
- the pulsation width ⁇ _rip is the amplitude of the waveform representing the time series change of the estimated frequency value ⁇ .
- the remuneration calculation data acquisition unit 42 obtains the pulsation width I rms _rip of the current effective value I rms from the acquired current effective value I rms .
- the pulsation width I rms _rip is the amplitude of the waveform representing the chronological change in the current effective value I rms .
- Each pulsation width V dc _rip, ⁇ _rip, and I rms _rip is an absolute value.
- the remuneration calculation data acquisition unit 42 uses an arbitrary filtering method such as a high-pass filter to obtain the pulsation widths Vdc_rip , ⁇ _rip , Extract I rms _rip respectively. Thereby, the remuneration calculation data acquisition unit 42 obtains the pulsation widths Vdc_rip , ⁇ _rip , and Irms_rip.
- the reward calculation data acquisition unit 42 stores the bus voltage detection value V dc and the pulsation widths V dc _rip, ⁇ _rip, and I rms _rip.
- the remuneration calculation data acquisition unit 42 stores the remuneration calculation data at preset timings when the synchronous machine 21 is in the driving state.
- the reward calculation data acquisition unit 42 acquires the bus voltage detection value V dc and the pulsation widths V dc _rip, ⁇ _rip, and I rms _rip at a plurality of timings, and stores the average value of the acquired data. Also good.
- V dc _drv V dc _rip_drv
- Irms_rip_drv Irms_rip_drv .
- the remuneration calculation data acquisition unit 42 acquires the frequency estimated value ⁇ , the current effective value I rms , and the bus voltage detection value V dc during the startup control period in which the synchronous machine 21 is in the free-run state.
- the remuneration calculation data acquisition unit 42 obtains the maximum value Vdc_rstmax of the bus voltage detection value Vdc during the activation control period.
- the remuneration calculation data acquisition unit 42 uses an arbitrary filtering method such as a high-pass filter to determine the maximum value Vdc_rip_rstmax of the pulsation width of the bus voltage detection value Vdc during the startup control period from the acquired bus voltage detection value Vdc .
- the reward calculation data acquisition unit 42 uses an arbitrary filtering method such as a high-pass filter to obtain the maximum pulsation width ⁇ _rip_rstmax of the frequency estimate ⁇ during the startup control period from the acquired frequency estimate ⁇ . .
- the remuneration calculation data acquisition unit 42 uses an arbitrary filtering method such as a high-pass filter to obtain the maximum value I rms _rip_rstmax of the pulsation width of the current rms value I rms during the startup control period from the acquired current rms value I rms. .
- the reward calculation data acquisition unit 42 obtains an increase V dc _rip_dat in the pulsation width V dc _rip, an increase ⁇ ⁇ _rip_dat in the pulsation width ⁇ ⁇ _rip, an increase I rms _rip_dat in the pulsation width I rms _rip, An increase V dc _dat of the bus voltage detection value V dc is obtained.
- the increments V dc _rip_dat, ⁇ _rip_dat, I rms _rip_dat, and V dc _dat are increments for when the synchronous machine 21 is in the driving state.
- the remuneration calculation data acquisition unit 42 obtains the increase amount V dc _rip_dat by the calculation shown in the following equation (15).
- Vdc_rip_dat Vdc_rip_rstmax ⁇ Vdc_rip_drv (15)
- the remuneration calculation data acquisition unit 42 obtains the increase amount ⁇ _rip_dat by the calculation shown in the following equation (16).
- ⁇ _rip_dat ⁇ _rip_rstmax ⁇ _rip_drv (16)
- the remuneration calculation data acquisition unit 42 obtains the increment I rms _rip_dat by the calculation shown in the following equation (17).
- Irms_rip_dat Irms_rip_rstmax ⁇ Irms_rip_drv (17)
- the remuneration calculation data acquisition unit 42 obtains the increase amount V dc — dat by the calculation shown in the following equation (18).
- Vdc_dat Vdc_rstmax ⁇ Vdc_drv (18)
- the remuneration calculation data acquisition unit 42 sends the increase amounts V dc _rip_dat, ⁇ _rip_dat, I rms _rip_dat, and V dc _dat, which are remuneration calculation data, to the model generation unit 43 .
- the model generation unit 43 generates a learned model using a data set created based on each value of the current control period T1 and the magnetic flux control period T2, the frequency estimation value ⁇ , and the current effective value Irms .
- the model generating unit 43 reduces the rise in the DC bus voltage in the power converter 11, reduces the pulsation width of the DC bus voltage, reduces the pulsation width of the frequency of the magnetic flux vector in the synchronous machine 21, Learning to infer the length of the current control period T1 and the length of the magnetic flux control period T2 from the frequency estimate ⁇ and the current rms value I rms Generate a finished model.
- the model generation unit 43 reduces the rise in the DC bus voltage, reduces the pulsation width of the DC bus voltage, reduces the pulsation width of the frequency of the magnetic flux vector, and reduces the pulsation width of the current effective value Irms . It is sufficient to generate a learned model for inferring the length of the current control period T1 and the length of the magnetic flux control period T2 that enable at least one of the following.
- the reward calculation data may be at least one of the amount of increase V dc _rip_dat, the amount of increase ⁇ _rip_dat, the amount of increase I rms _rip_dat, and the amount of increase V dc _dat.
- the reward calculation data When aiming to reduce the pulsation width of the DC bus voltage, the reward calculation data includes the increment V dc _rip_dat. In order to reduce the pulsation width of the frequency of the magnetic flux vector, the reward calculation data includes an increase amount ⁇ _rip_dat. When reducing the pulsation width of the current effective value I rms , the reward calculation data includes the increase I rms _rip_dat. When aiming to reduce the amount of increase in the DC bus voltage, the remuneration calculation data includes the amount of increase V dc _dat.
- supervised learning is that an action subject, an agent, in an environment observes the current state and decides what action to take. Agents obtain rewards from the environment by selecting actions, and learn policies that maximize rewards through a series of actions.
- Q-learning and TD-learning are known.
- the action-value table which is a general update formula for the action-value function Q(s, a)
- the action value function Q(s, a) represents the action value Q, which is the action value of selecting the action 'a' under the environment 's'.
- 's t ' represents the state of the environment at time 't'.
- "a t " represents an action at time “t”.
- Action "a t " changes the state from “s t " to "s t+1 ".
- 'r t+1 ' represents the reward obtained by changing the state from 's t ' to 's t+1 '.
- “ ⁇ ” represents a discount rate and satisfies 0 ⁇ 1.
- “ ⁇ ” represents a learning coefficient and satisfies 0 ⁇ 1.
- the action "a t " is each value of the current control period T1 and the magnetic flux control period T2.
- the state 's t ' is the frequency estimate ⁇ and the current rms I rms .
- the model generator 43 learns the best action 'a t ' in the state 's t ' at the time 't'.
- the update formula represented by the above formula (19) is that if the action value of the best action “a” at time “t+1" is greater than the action value Q of the action "a” executed at time “t” , the action value Q is increased, and vice versa, the action value Q is decreased.
- the action value function Q(s, a) is updated so that the action value Q of action "a” at time “t” approaches the best action value at time "t+1".
- the model generator 43 has a reward calculator 44 and a function updater 45 . Based on the increments Vdc_rip_dat , ⁇ _rip_dat , Irms_rip_dat , and Vdc_dat, the reward calculation unit 44 calculates each value of the current control period T1 and the magnetic flux control period T2, the estimated frequency value ⁇ , and the effective current Calculate the reward for the combination with the value I rms .
- the function updating unit 45 updates the function for obtaining the respective values of the current control period T1 and the magnetic flux control period T2 from the estimated frequency value ⁇ and the effective current value I rms according to the reward.
- the function update unit 45 outputs the learned model created by updating the function to the learned model storage unit 46 .
- the learned model storage unit 46 stores learned models.
- the reward calculation unit 44 calculates each value of the current control period T1 and the magnetic flux control period T2, the estimated frequency value ⁇ , the current effective value I rms , and the amount of increase V dc _rip_dat, ⁇ ⁇ _rip_dat, I rms _rip_dat, V dc Calculate the reward 'r' based on _dat. For example, when at least one of the increments V dc _rip_dat, ⁇ _rip_dat, I rms _rip_dat, and V dc _dat decreases as a result of changing the value of the current control period T1 or the value of the magnetic flux control period T2, the reward calculation unit 44 increases the reward "r". The reward calculation unit 44 increases the reward "r" by giving "1", which is the value of the reward. Note that the reward value is not limited to "1".
- the reward calculation unit 44 decreases the reward "r".
- the reward calculation unit 44 reduces the reward "r” by giving "-1", which is the value of the reward. Note that the reward value is not limited to "-1".
- the reward calculation data can be selected according to the installation environment of the synchronous machine 21 and the like.
- Each of the increments V dc _rip_dat, ⁇ _rip_dat, and I rms _rip_dat may be omitted as appropriate when the pulsation width is sufficiently below the system requirements and no measures for reduction are necessary.
- the function update unit 45 updates the function for determining each value of the current control period T1 and the magnetic flux control period T2 according to the reward calculated by the reward calculation unit 44.
- the machine learning device 40 repeatedly executes the learning described above.
- the function update unit 45 outputs the learned model created by updating the function to the learned model storage unit 46 .
- the machine learning device 40 determines each value of the current control period T1 and the magnetic flux control period T2 for the action value function Q(s t , at ) represented by the above equation (19). It is used as a function for
- the learned model storage unit 46 stores the action-value function Q(s t , a t ) updated by the function update unit 45, that is, the learned model.
- FIG. 9 is a flow chart showing the processing procedure of the machine learning device 40 according to the fourth embodiment.
- a reinforcement learning method for updating the action-value function Q(s, a) will be described with reference to the flowchart of FIG.
- step S1 the machine learning device 40 acquires values of the current control period T1 and the magnetic flux control period T2, the estimated frequency value ⁇ , and the effective current value I rms in the learning data acquisition unit 41 . That is, the machine learning device 40 acquires learning data. In addition, the machine learning device 40 acquires the increments Vdc_rip_dat , ⁇ _rip_dat , Irms_rip_dat , and Vdc_dat, which are data for remuneration calculation, by calculation in the remuneration calculation data acquisition unit 42 .
- step S ⁇ b>2 the machine learning device 40 uses the reward calculator 44 to calculate the reward.
- the reward calculator 44 calculates a reward for a combination of each value of the current control period T1 and the magnetic flux control period T2, the estimated frequency value ⁇ , and the effective current value Irms .
- the remuneration calculation unit 44 increases or decreases the remuneration based on the comparison result between the acquired remuneration calculation data and the previous remuneration calculation data.
- the machine learning device 40 updates the action-value function Q(s, a) based on the reward calculated at step S2.
- the machine learning device 40 updates the action-value function Q(s t , at ) stored in the learned model storage unit 46 .
- step S4 the machine learning device 40 determines whether or not the action value function Q(s, a) has converged.
- the machine learning device 40 determines that the action-value function Q(s, a) has converged by not updating the action-value function Q(s, a) in step S3.
- step S4 When it is determined that the action-value function Q(s, a) has not converged (step S4, No), the machine learning device 40 returns the procedure to step S1. When it is determined that the action-value function Q(s, a) has converged (step S4, Yes), the machine learning device 40 terminates learning according to the procedure shown in FIG. Note that the machine learning device 40 may continue learning by returning the procedure from step S3 to step S1 without making the determination in step S4.
- the learned model storage unit 46 stores a learned model that is the generated action-value function Q(s, a).
- reinforcement learning is applied to the learning algorithm used by the machine learning device 40
- learning other than reinforcement learning may be applied to the learning algorithm.
- the machine learning device 40 executes machine learning using known learning algorithms other than reinforcement learning, such as deep learning, neural networks, genetic programming, inductive logic programming, or support vector machines. Also good.
- the machine learning device 40 shown in FIG. 8 is provided outside the rotating machine control device 20A.
- the machine learning device 40 may be a device built into the rotating machine control device 20A.
- the machine learning device 40 may be a device connectable to the rotating machine control device 20A via a network.
- Machine learning device 40 may be a device that resides on a cloud server.
- the machine learning device 40 may learn each value of the current control period T1 and the magnetic flux control period T2 according to a data set created for a plurality of rotating machine control devices 20A.
- the machine learning device 40 may acquire learning data from a plurality of rotating machine control devices 20A used at the same location, or acquire learning data from a plurality of rotating machine control devices 20A used at different locations. You can get the data.
- the learning data may be collected from a plurality of rotating machine control devices 20A that operate independently of each other at a plurality of locations. After starting to collect learning data from a plurality of rotating machine control devices 20A, a new rotating machine control device 20A may be added as a target for which learning data is collected. Further, after starting to collect learning data from the plurality of rotating machine control devices 20A, some of the plurality of rotating machine control devices 20A may be excluded from targets for which learning data is collected.
- the machine learning device 40 that has learned about one rotating machine control device 20A may also learn about other rotating machine control devices 20A other than the rotating machine control device 20A.
- the machine learning device 40 that performs learning for the other rotating machine control device 20A can update the learned model by re-learning in the other rotating machine control device 20A.
- the machine learning device 40 generates a trained model for inferring the values of the current control period T1 and the magnetic flux control period T2 from the frequency estimate ⁇ and the current effective value I rms .
- the machine learning device 40 reduces the rise in the DC bus voltage in the power converter 11, reduces the pulsation width of the DC bus voltage, reduces the pulsation width of the frequency of the magnetic flux vector in the synchronous machine 21, and reduces the pulsation width in the synchronous machine 21. It is possible to generate a learned model for obtaining the values of the current control period T1 and the magnetic flux control period T2 that enable reduction of the pulsation width of the current effective value I rms of the flowing current.
- Embodiment 5 In a fifth embodiment, an inference device that infers the length of the current control period T1 and the length of the magnetic flux control period T2 using the learned model generated by the machine learning apparatus 40 according to the fourth embodiment will be described.
- FIG. 10 is a diagram showing an inference device 50 and a rotary machine control device 20 according to the fifth embodiment.
- Embodiment 5 the same components as those in Embodiments 1 to 4 are denoted by the same reference numerals, and configurations different from those in Embodiments 1 to 4 will be mainly described.
- the inference device 50 shown in FIG. 10 is provided outside the rotating machine control device 20 . 10 shows only the learned model storage unit 46 of the machine learning device 40. As shown in FIG.
- the rotating machine control device 20 is the rotating machine control device 20 according to any one of the first to third embodiments.
- the rotating machine control device 20A of the fourth embodiment may be applied instead of the rotating machine control device 20 of any one of the first to third embodiments.
- the inference device 50 reduces the rise of the DC bus voltage in the power converter 11, reduces the pulsation width of the DC bus voltage, and reduces the pulsation width of the frequency of the magnetic flux vector in the synchronous machine 21, for the rotating machine control device 20. , and the length of the current control period T1 and the length of the magnetic flux control period T2 that make it possible to reduce the pulsation width of the current effective value of the current flowing through the synchronous machine 21 are inferred.
- the inference device 50 has an inference data acquisition unit 51 and an inference unit 52 .
- the inference data acquisition unit 51 acquires inference data, which are the estimated frequency value ⁇ and the current effective value I rms , from the voltage commander 10 of the rotating machine control device 20 .
- the inference data acquisition unit 51 sends the acquired inference data to the inference unit 52 .
- the inference unit 52 reads the learned model generated by the machine learning device 40 according to the fourth embodiment from the learned model storage unit 46 .
- the inference unit 52 infers the length of the current control period T1 and the length of the magnetic flux control period T2 by inputting the estimated frequency value ⁇ and the effective current value Irms , which are data for inference, to the learned model.
- Each value of the current control period T1 and the magnetic flux control period T2, which are inference results, are output from the learned model.
- the inference unit 52 sends the values of the current control period T1 and the magnetic flux control period T2 output from the learned model to the rotating machine control device 20 .
- FIG. 11 is a flowchart of the processing procedure of the inference device 50 according to the fifth embodiment.
- the inference device 50 acquires the frequency estimation value ⁇ and current effective value I rms as inference data in the inference data acquisition unit 51 .
- the inference unit 52 of the inference unit 52 inputs the estimated frequency value ⁇ and the effective current value I rms to the trained model, and obtains the length of the current control period T1 and the length of the magnetic flux control period T2.
- step S ⁇ b>13 the inference device 50 outputs each value of the current control period T ⁇ b>1 and the magnetic flux control period T ⁇ b>2 from the inference unit 52 to the rotating machine control device 20 .
- the inference device 50 ends the processing according to the procedure shown in FIG.
- the inference device 50 shown in FIG. 10 is provided outside the rotary machine control device 20.
- the inference device 50 may be a device built into the rotating machine control device 20 .
- the inference device 50 may be a device connectable to the rotating machine control device 20 via a network.
- the inference device 50 may be a device existing on a cloud server.
- the inference device 50 infers the length of the current control period T1 and the length of the magnetic flux control period T2 from the learned model based on the learning data acquired from the rotating machine control device 20A, and performs current control.
- Each value of the period T1 and the magnetic flux control period T2 is sent to the rotary machine control device 20 other than the rotary machine control device 20A.
- the rotary machine controllers 20 and 20A to which the values of the current control period T1 and the magnetic flux control period T2 are sent may be the rotary machine controllers 20 and 20A from which the learning data is acquired.
- the rotary machine control devices 20 and 20A other than the rotary machine control devices 20 and 20A may be used.
- Inference device 50 performs inference using a learned model to reduce the amount of rise in the DC bus voltage and to reduce the pulsations of the DC bus voltage, frequency, and current effective value Irms .
- the length and length of the flux control period T2 can be inferred.
- the rotating machine control device 20 receives the values of the current control period T1 and the flux control period T2 sent by the reasoning device 50 .
- the rotating machine control device 20 determines the length of the current control period T1 and the length of the magnetic flux control period T2 in the startup control period based on the respective values of the current control period T1 and the magnetic flux control period T2.
- Rotating machine control device 20 reduces the amount of rise in the DC bus voltage in power converter 11, reduces the pulsation width of the DC bus voltage, reduces the pulsation width of the frequency of the magnetic flux vector in synchronous machine 21, It is possible to reduce the pulsation width of the current effective value I rms of the current flowing through.
- the rotating machine control device 20 is able to recover the synchronous machine 21 from the free-run state in a state in which the rise in the DC bus voltage is reduced and the pulsations of the DC bus voltage, the frequency, and the current effective value Irms are reduced. can be started stably.
- the rotating machine control device 20 can stabilize the startup of the synchronous machine 21 .
- the rotating machine control device 20 can reduce the generation of unintended torque by being able to reduce fluctuations in the DC bus voltage.
- the rotary machine control device 20 can start the synchronous machine 21 from the free-run state in a state in which the generation of unintended torque is reduced.
- each of the learning data and the inference data includes the bus voltage detection value Vdc .
- FIG. 12 is a diagram showing a machine learning device 40A and a rotating machine control device 20A according to a modification of the fifth embodiment.
- FIG. 13 is a diagram showing an inference device 50A and a rotating machine control device 20 according to a modification of the fifth embodiment.
- the learning data acquisition unit 41A of the machine learning device 40A shown in FIG. and the current effective value I rms .
- the learning data acquisition unit 41A further acquires the bus voltage detection value V dc from the voltage detector 15 .
- the learning data acquired by the learning data acquisition unit 41A includes values of the current control period T1 and the magnetic flux control period T2, the estimated frequency value ⁇ , the current effective value I rms , and the bus voltage detection value V dc . include.
- the learning data acquisition unit 41A obtains a data set in which each value of the current control period T1 and the magnetic flux control period T2, the estimated frequency value ⁇ , the current effective value I rms , and the bus voltage detection value V dc are put together. create.
- the model generation unit 43 generates a data set based on each value of the current control period T1 and the magnetic flux control period T2, the estimated frequency value ⁇ , the current effective value Irms , and the bus voltage detection value Vdc . Generate a trained model using Thereby, the machine learning device 40A generates a learned model for inferring the respective values of the current control period T1 and the magnetic flux control period T2 from the estimated frequency value ⁇ , the effective current value I rms and the bus voltage detection value V dc . do.
- An inference data acquisition unit 51A of an inference device 50A shown in FIG. 13 acquires a frequency estimated value ⁇ and a current effective value I rms from the rotating machine control device 20, similarly to the inference data acquisition unit 51 shown in FIG. do.
- the inference data acquisition unit 51A further acquires the bus voltage detection value Vdc from the voltage detector 15 of the rotating machine control device 20 .
- the inference data acquired by the inference data acquisition unit 51A includes the estimated frequency value ⁇ , the current effective value I rms , and the bus voltage detection value V dc .
- the inference unit 52 uses the learned model generated by the machine learning device 40A to determine the length of the current control period T1 and the magnetic flux control period T2 from the estimated frequency value ⁇ , the effective current value Irms , and the bus voltage detection value Vdc . infer the length of The inference unit 52 inputs the estimated frequency value ⁇ , the effective current value I rms , and the detected bus voltage V dc , which are data for inference, to the learned model, thereby determining the length of the current control period T1 and the magnetic flux control period T2. infer the length of
- machine learning device 40A can generate a more accurate learned model by adding bus voltage detection value Vdc to learning data.
- Inference device 50A by reasoning using a learned model, reduces the rise in the DC bus voltage and reduces the pulsations of the DC bus voltage, frequency, and current effective value Irms .
- a highly accurate inference can be made about the length and the length of the magnetic flux control period T2.
- the machine learning devices 40, 40A according to the fourth or fifth embodiment are implemented by a hardware configuration similar to the configuration shown in FIG.
- the learning data acquisition units 41 and 41A, the reward calculation data acquisition unit 42, and the model generation unit 43 are implemented by a processing circuit 61 having a processor 63 and a memory 64.
- the learning data acquisition units 41 and 41A, the reward calculation data acquisition unit 42, and the model generation unit 43 may be realized by dedicated processing circuits.
- the reasoning devices 50 and 50A according to the fifth embodiment are implemented by a hardware configuration similar to the configuration shown in FIG.
- the inference data acquisition units 51 and 51A and the inference unit 52 are implemented by a processing circuit 61 having a processor 63 and a memory 64 .
- the inference data acquisition units 51 and 51A and the inference unit 52 may be realized by dedicated processing circuits.
- the machine learning devices 40 and 40A may include the amount of increase in the torque detection value of the synchronous machine 21 in the reward calculation data.
- a reward calculation data acquisition unit 42 of the machine learning devices 40 and 40A acquires a torque detection value from a torque detector installed in the synchronous machine 21 .
- the remuneration calculation data acquisition unit 42 obtains the amount of increase in the torque detection value during the activation control period.
- the amount of increase is the amount of increase when the synchronous machine 21 is in the driving state.
- the remuneration calculator 44 increases the remuneration when the amount of increase in the detected torque value decreases, and decreases the value of the remuneration when the amount of increase in the detected torque value increases.
- the machine learning devices 40 and 40A can generate learned models for obtaining values of the current control period T1 and the magnetic flux control period T2 that can reduce unintended torque increases.
- a torque detection value may be added to each of the learning data and the inference data.
- Learning data acquisition units 41 and 41A of machine learning devices 40 and 40A acquire torque detection values from torque detectors.
- the inference data acquisition units 51 and 51A of the inference devices 50 and 50A acquire torque detection values from the torque detectors.
- the machine learning devices 40 and 40A can generate a more accurate learned model by adding the torque detection value to the learning data.
- the inference devices 50, 50A use the learned models generated by the machine learning devices 40, 40A to infer the length of the current control period T1 and the length of the magnetic flux control period T2 that can reduce the increase in torque. be able to.
- each embodiment is an example of the content of the present disclosure.
- the configuration of each embodiment can be combined with another known technique. Configurations of respective embodiments may be combined as appropriate. A part of the configuration of each embodiment can be omitted or changed without departing from the gist of the present disclosure.
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Control Of Ac Motors In General (AREA)
Abstract
Description
図1は、実施の形態1にかかる回転機制御装置20の構成例を示す図である。回転機制御装置20は、交流回転機である同期機21を制御する。以下の説明において、固定座標系における二軸をα軸およびβ軸、回転座標系における二軸をd軸およびq軸と称する。実施の形態1では、一次磁束ベクトルの位相に制御位相を同期させる場合について説明する。一次磁束ベクトルの位相を用いて、一次磁束ベクトルの方向を基準のd軸とする回転座標への座標変換を行った場合におけるd軸およびq軸を、それぞれds軸、qs軸と表す。ds軸は、一次磁束ベクトルの方向の軸である。qs軸は、一次磁束ベクトルに直交する方向の軸である。
実施の形態1では、d軸の一次磁束およびq軸の一次磁束から磁束推定値|φ^|を求める例について説明した。実施の形態2では、α軸の一次磁束およびβ軸の一次磁束から磁束推定値|φ^|を求める例について説明する。
実施の形態1および2では、一次磁束から磁束推定値|φ^|を求める例について説明した。実施の形態3では、二次磁束から磁束推定値|φ^|を求める例について説明する。
実施の形態4では、電流制御期間T1の長さと磁束制御期間T2の長さとを推論するための学習済モデルを生成する機械学習装置について説明する。
Vdc_rip_dat=Vdc_rip_rstmax-Vdc_rip_drv ・・・(15)
ω^_rip_dat=ω^_rip_rstmax-ω^_rip_drv ・・・(16)
Irms_rip_dat=Irms_rip_rstmax-Irms_rip_drv ・・・(17)
Vdc_dat=Vdc_rstmax-Vdc_drv ・・・(18)
実施の形態5では、実施の形態4にかかる機械学習装置40によって生成された学習済モデルを用いて電流制御期間T1の長さと磁束制御期間T2の長さとを推論する推論装置について説明する。
Claims (12)
- 回転機に流れる交流電流を検出して電流検出値を出力する電流検出器と、
電圧指令値に基づいた交流電圧の印加によって前記回転機へ電力を供給する電力変換器と、
前記電流検出値が電流指令値に一致するように前記電圧指令値を調整する電流制御器と、
前記回転機における磁束ベクトルの振幅の推定値である磁束推定値を求める推定器と、
前記回転機が、前記電力変換器による電力供給が遮断されて惰性で回転する状態となってから前記電力供給が再開されるまでの起動制御期間において、設定された磁束指令値に前記磁束推定値が一致するように前記電流指令値を調整する磁束制御器と、
を備えることを特徴とする回転機制御装置。 - 前記磁束制御器が出力する前記電流指令値は、前記回転機の回転座標系における二軸のうちの1つであるd軸の電流指令値であることを特徴とする請求項1に記載の回転機制御装置。
- 前記磁束ベクトルは、前記回転機の一次磁束のベクトル、または、前記回転機の二次磁束のベクトルであることを特徴とする請求項1または2に記載の回転機制御装置。
- 前記起動制御期間には、前記磁束制御器が出力する前記電流指令値を基に電流制御が行われる磁束制御期間の前に、あらかじめ設定された電流指令値である既定指令値を基に電流制御が行われる電流制御期間が含まれることを特徴とする請求項1から3のいずれか1つに記載の回転機制御装置。
- 前記起動制御期間において、前記磁束制御期間は前記電流制御期間よりも長いことを特徴とする請求項4に記載の回転機制御装置。
- 前記磁束ベクトルの周波数推定値と前記電流検出値から求まる電流実効値とを取得する推論用データ取得部と、
機械学習による学習済モデルへ前記周波数推定値と前記電流実効値とを入力することによって、前記電流制御期間の長さと前記磁束制御期間の長さとを推論する推論部と、
をさらに備えることを特徴とする請求項4または5に記載の回転機制御装置。 - 前記電流制御期間の長さを示す値と前記磁束制御期間の長さを示す値と前記周波数推定値と前記電流実効値とを取得する学習用データ取得部と、
前記電流制御期間の長さを示す値と前記磁束制御期間の長さを示す値と前記周波数推定値と前記電流実効値とに基づいて作成されたデータセットを用いて前記学習済モデルを生成するモデル生成部と、
をさらに備えることを特徴とする請求項6に記載の回転機制御装置。 - 請求項4または5に記載の回転機制御装置について、電力変換器における直流母線電圧の上昇分の低減と、前記直流母線電圧の脈動幅の低減と、回転機における磁束ベクトルの周波数の脈動幅の低減と、前記回転機に流れる電流の電流実効値の脈動幅の低減との少なくとも1つを可能とする電流制御期間の長さと磁束制御期間の長さとを学習する機械学習装置であって、
前記電流制御期間の長さを示す値と前記磁束制御期間の長さを示す値と前記磁束ベクトルの周波数推定値と前記電流実効値とを取得する学習用データ取得部と、
前記電流制御期間の長さを示す値と前記磁束制御期間の長さを示す値と前記周波数推定値と前記電流実効値とに基づいて作成されたデータセットを用いて学習済モデルを生成するモデル生成部と、
前記学習済モデルを記憶する学習済モデル記憶部と、
を備えることを特徴とする機械学習装置。 - 前記直流母線電圧を検出して母線電圧検出値を出力する電圧検出器と、
前記母線電圧検出値と前記周波数推定値と前記電流実効値とが入力され、前記母線電圧検出値の脈動幅の増加量と、前記周波数推定値の脈動幅の増加量と、前記電流実効値の脈動幅の増加量と、前記母線電圧検出値の増加量との少なくとも1つである報酬計算用データを求める報酬計算用データ取得部と、
をさらに備え、
前記モデル生成部は、
前記報酬計算用データを基に、前記電流制御期間の長さを示す値と前記磁束制御期間の長さを示す値と前記周波数推定値と前記電流実効値との組み合わせに対する報酬を計算する報酬計算部と、
前記周波数推定値および前記電流実効値から前記電流制御期間の長さを示す値と前記磁束制御期間の長さを示す値とを求めるための関数を前記報酬に従って更新する関数更新部と、を備え、
前記関数更新部は、前記関数の更新によって生成された前記学習済モデルを出力することを特徴とする請求項8に記載の機械学習装置。 - 前記学習用データ取得部は、さらに、前記母線電圧検出値を取得し、
前記モデル生成部は、前記電流制御期間の長さを示す値と前記磁束制御期間の長さを示す値と前記周波数推定値と前記電流実効値と前記母線電圧検出値とに基づいて作成された前記データセットを用いて前記学習済モデルを生成することを特徴とする請求項9に記載の機械学習装置。 - 請求項4または5に記載の回転機制御装置について、電力変換器における直流母線電圧の上昇分の低減と、前記直流母線電圧の脈動幅の低減と、回転機における磁束ベクトルの周波数の脈動幅の低減と、前記回転機に流れる電流の電流実効値の脈動幅の低減との少なくとも1つを可能とする電流制御期間の長さと磁束制御期間の長さとを推論する推論装置であって、
前記磁束ベクトルの周波数推定値と、前記電流検出値から求まる電流実効値とを取得する推論用データ取得部と、
前記周波数推定値および前記電流実効値から前記電流制御期間の長さと前記磁束制御期間の長さとを推論するための学習済モデルへ前記周波数推定値および前記電流実効値を入力することによって、前記電流制御期間の長さと前記磁束制御期間の長さとを推論する推論部と、
を備えることを特徴とする推論装置。 - 前記推論用データ取得部は、さらに、前記直流母線電圧の検出値である母線電圧検出値を取得し、
前記推論部は、前記周波数推定値、前記電流実効値および前記母線電圧検出値から前記電流制御期間の長さと前記磁束制御期間の長さとを推論するための学習済モデルを用いて、前記周波数推定値、前記電流実効値および前記母線電圧検出値から前記電流制御期間の長さと前記磁束制御期間の長さとを推論することを特徴とする請求項11に記載の推論装置。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2021567903A JP7038931B1 (ja) | 2021-05-20 | 2021-05-20 | 回転機制御装置、機械学習装置および推論装置 |
CN202180086576.2A CN116648848A (zh) | 2021-05-20 | 2021-05-20 | 旋转机控制装置、机器学习装置及推断装置 |
US18/032,569 US11929694B2 (en) | 2021-05-20 | 2021-05-20 | Rotary machine control apparatus, machine learning apparatus, and inference apparatus |
PCT/JP2021/019223 WO2022244204A1 (ja) | 2021-05-20 | 2021-05-20 | 回転機制御装置、機械学習装置および推論装置 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2021/019223 WO2022244204A1 (ja) | 2021-05-20 | 2021-05-20 | 回転機制御装置、機械学習装置および推論装置 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022244204A1 true WO2022244204A1 (ja) | 2022-11-24 |
Family
ID=81213700
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2021/019223 WO2022244204A1 (ja) | 2021-05-20 | 2021-05-20 | 回転機制御装置、機械学習装置および推論装置 |
Country Status (4)
Country | Link |
---|---|
US (1) | US11929694B2 (ja) |
JP (1) | JP7038931B1 (ja) |
CN (1) | CN116648848A (ja) |
WO (1) | WO2022244204A1 (ja) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2009171712A (ja) * | 2008-01-15 | 2009-07-30 | Fuji Electric Systems Co Ltd | 速度センサレスベクトル制御装置 |
JP2010114969A (ja) * | 2008-11-05 | 2010-05-20 | Mitsubishi Electric Corp | 電力変換装置 |
JP2011244655A (ja) * | 2010-05-20 | 2011-12-01 | Toshiba Corp | 回転センサレス制御装置 |
JP2018007390A (ja) * | 2016-06-30 | 2018-01-11 | シンフォニアテクノロジー株式会社 | モータ制御装置 |
JP2018099025A (ja) * | 2016-12-15 | 2018-06-21 | 日本油圧工業株式会社 | 交流電動機の始動と運転のための装置及び方法 |
Family Cites Families (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4437629B2 (ja) * | 2001-06-14 | 2010-03-24 | 三菱電機株式会社 | 電気車制御装置 |
JP2003284381A (ja) * | 2002-03-26 | 2003-10-03 | Sumitomo Heavy Ind Ltd | 誘導電動機の瞬停再起動方法及びインバータ制御装置 |
JP2004040837A (ja) | 2002-06-28 | 2004-02-05 | Meidensha Corp | Pmモータの位置・速度センサレス制御装置 |
JP3719426B2 (ja) * | 2002-07-08 | 2005-11-24 | 株式会社安川電機 | 交流電動機の制御方法及び制御装置 |
EP2034604A4 (en) * | 2006-06-29 | 2012-11-14 | Mitsubishi Electric Corp | AC ROTARY MACHINE CONTROL |
WO2008004294A1 (fr) * | 2006-07-06 | 2008-01-10 | Mitsubishi Electric Corporation | Dispositif de commande de vecteur de moteur à induction, procédé de commande de vecteur de moteur à induction, et dispositif de commande d'entraînement de moteur à induction |
JP2009165281A (ja) * | 2008-01-08 | 2009-07-23 | Fuji Electric Systems Co Ltd | 速度センサレスベクトル制御装置 |
CA2806317C (en) | 2010-07-27 | 2015-08-11 | Mitsubishi Electric Corporation | Control apparatus for ac rotating machine |
ES2729375T3 (es) * | 2010-07-28 | 2019-11-04 | Mitsubishi Electric Corp | Aparato de control para una máquina rotatoria en CA |
JP6447183B2 (ja) * | 2015-01-30 | 2019-01-09 | 富士電機株式会社 | 誘導電動機の制御装置 |
JP6556403B2 (ja) * | 2017-03-10 | 2019-08-07 | 三菱電機株式会社 | 電気車推進制御装置 |
JP2021040353A (ja) * | 2017-11-08 | 2021-03-11 | 株式会社日立製作所 | 誘導電動機の駆動装置および駆動方法 |
US11837982B2 (en) * | 2021-04-28 | 2023-12-05 | Panasonic Intellectual Property Management Co., Ltd. | Rotary machine control device |
-
2021
- 2021-05-20 US US18/032,569 patent/US11929694B2/en active Active
- 2021-05-20 WO PCT/JP2021/019223 patent/WO2022244204A1/ja active Application Filing
- 2021-05-20 CN CN202180086576.2A patent/CN116648848A/zh active Pending
- 2021-05-20 JP JP2021567903A patent/JP7038931B1/ja active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2009171712A (ja) * | 2008-01-15 | 2009-07-30 | Fuji Electric Systems Co Ltd | 速度センサレスベクトル制御装置 |
JP2010114969A (ja) * | 2008-11-05 | 2010-05-20 | Mitsubishi Electric Corp | 電力変換装置 |
JP2011244655A (ja) * | 2010-05-20 | 2011-12-01 | Toshiba Corp | 回転センサレス制御装置 |
JP2018007390A (ja) * | 2016-06-30 | 2018-01-11 | シンフォニアテクノロジー株式会社 | モータ制御装置 |
JP2018099025A (ja) * | 2016-12-15 | 2018-06-21 | 日本油圧工業株式会社 | 交流電動機の始動と運転のための装置及び方法 |
Also Published As
Publication number | Publication date |
---|---|
JP7038931B1 (ja) | 2022-03-18 |
US20230353075A1 (en) | 2023-11-02 |
US11929694B2 (en) | 2024-03-12 |
CN116648848A (zh) | 2023-08-25 |
JPWO2022244204A1 (ja) | 2022-11-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Marino et al. | On-line stator and rotor resistance estimation for induction motors | |
JP4989075B2 (ja) | 電動機駆動制御装置及び電動機駆動システム | |
JP4411796B2 (ja) | 速度センサを持たない誘導モータドライブの制御システム、オブザーバ及び制御方法 | |
TW201830846A (zh) | 用於啟動同步馬達的系統及方法 | |
JP5073063B2 (ja) | 永久磁石同期電動機の制御装置 | |
CN1741367B (zh) | 电动机的无位置传感器控制电路 | |
EP1986317A1 (en) | Stator resistance adaptation in sensorless PMSM drives | |
JP2004166408A (ja) | 永久磁石同期電動機制御方法 | |
JP2008511274A (ja) | 誘導機の電力調整方法および装置 | |
JP2006191737A (ja) | モータ制御装置及びこれを有するモータ駆動システム | |
CN110557075A (zh) | 电机转矩的确定方法及装置、电机控制系统 | |
JP6485810B2 (ja) | 誘導モーターのローター角速度を制御するための方法及びシステム | |
CN110943657A (zh) | 永磁同步电机无模型自适应转速控制方法及系统 | |
JP2002136197A (ja) | センサレスベクトル制御装置およびその方法 | |
JP6851544B2 (ja) | 誘導モータの動作を制御するモータ駆動装置及び方法 | |
WO2022244204A1 (ja) | 回転機制御装置、機械学習装置および推論装置 | |
Sayouti et al. | Sensor less low speed control with ANN MRAS for direct torque controlled induction motor drive | |
JP5499594B2 (ja) | 永久磁石形同期電動機の制御装置 | |
Lian et al. | Robust adaptive control of linear induction motors with unknown end-effect and secondary resistance | |
JP5334524B2 (ja) | 永久磁石同期モータの制御装置及びその制御方法 | |
Himker et al. | A family of adaptive position estimators for pmsm using the gradient descent method | |
JP2018014842A (ja) | 自動測定システム及び自動測定方法 | |
JP4667741B2 (ja) | 誘導電動機の制御装置 | |
Younesi et al. | Design of a sensorless controller for PMSM using Krill Herd algorithm | |
Chehimi et al. | A predictive controller for induction motors using an unknown input observer for online rotor resistance estimation and an update technique for load torque |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
ENP | Entry into the national phase |
Ref document number: 2021567903 Country of ref document: JP Kind code of ref document: A |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21940818 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 202180086576.2 Country of ref document: CN |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 21940818 Country of ref document: EP Kind code of ref document: A1 |