Driving control device and driving control method for induction motor

Abstract

In order to make it possible to instantaneously estimate the RPM for a wide range of RPMs without using a rotation speed sensor with respect to an induction motor that idles in a non-energized state, and to smoothly transition the induction motor into operation, this drive control device for an induction motor comprises an inverter for driving the induction motor, an electric current detector for detecting a primary electric current of the induction motor as a detected primary electric current, a self-excited oscillator for generating a voltage command for the inverter and self-oscillating the primary electric current of the induction motor, and an initial speed estimator for estimating the initial speed of the induction motor on the basis of the self-excited detected primary electric current, the induction motor furthermore having a voltage corrector for superimposing an additional voltage command onto the voltage command, and the initial speed estimator estimating the initial speed in a state in which the voltage corrector superimposes the additional voltage command onto the voltage command.

Description

Induction motor drive control device and drive control method

[0001] The present invention relates to a drive control device and a drive control method for an induction motor for driving the induction motor at a variable speed without using a rotational speed sensor.

[0002] The technology of variable speed drive of induction motors without using a rotational speed sensor is widely used in industrial fields and drive systems for railway electric vehicles because it can reduce costs and maintenance.Instead of using a rotational speed sensor, the rotational speed of the induction motor is estimated by calculation within the controller.This speed estimation is performed using a method that calculates the speed electromotive voltage (back electromotive force) generated inside the induction motor due to rotation.

[0003] However, since induction motors do not generate speed electromotive force due to rotation unless an excitation current is passed through them, information about the rotation speed cannot be obtained when the induction motor is in a state of idling without current being supplied. Therefore, in order to restart an induction motor from idling in a short time, some ingenuity is required to obtain information about the rotation speed.

[0004] Patent Document 1 and Non-Patent Document 1 disclose a method for estimating the rotation speed during idling by utilizing the characteristic that the impedance characteristic of an induction motor becomes high at high rotation speeds. In this method, an impulse voltage is applied to an induction motor in an idling state, and then the current generated is positively fed back to cause an oscillation phenomenon, and the rotation speed of the induction motor is estimated and calculated from the oscillation frequency.

[0005] Patent Document 2 discloses a method in which an AC voltage of a predetermined frequency is applied to an induction motor while it is idling, the frequency of the applied AC is corrected so that the current component that contributes to torque among the current components generated at that time becomes zero, and this applied frequency is adjusted to the same frequency as the rotation speed.

[0006] Patent Document 3 discloses a method in which a DC current is passed through an induction motor during idling, the secondary magnetic flux component of the induction motor is estimated and calculated, the gradient of the differential value is calculated, and the idling speed is estimated from the average value of the calculated gradient.

[0007] Japanese Patent Laid-Open No. 11-346500 Japanese Patent Laid-Open No. 2001-238497 Japanese Patent Laid-Open No. 2018-011474

[0008] "Restart Method for Induction Motor Drive System without Speed ​​Sensor", IEEJ Transactions on Electrical Engineering, Vol. 119, No. 2, pp. 211-216, 1999

[0009] The methods disclosed in Patent Document 1 and Non-Patent Document 1 may not be able to achieve stable self-excited vibration when the secondary time constant of the induction motor is long. Furthermore, as a result of positive feedback, the flowing current may contain a DC component, making it difficult to accurately measure the vibration frequency. To avoid these issues, Patent Document 1 proposes the use of a high-pass filter, but using a high-pass filter makes it difficult to estimate speed at low speeds. Furthermore, the current control system for positive feedback requires the actual current to follow the AC current command. Therefore, the higher the drive frequency, the higher the current control gain must be set to ensure continued vibration. Therefore, it is difficult to apply this method to high-speed induction motors.

[0010] The method disclosed in Patent Document 2 applies an AC current of a predetermined frequency to gradually approach the actual rotation frequency, but if the speed difference (frequency difference) between the two is large, it takes a long time to estimate the actual rotation frequency. Therefore, depending on the conditions, it may be difficult to estimate the initial speed in a short time. Furthermore, depending on the constants of the induction motor, it may be difficult for the excitation current to flow, making it difficult to estimate the initial rotation frequency.

[0011] The method disclosed in Patent Document 3 requires estimating the secondary magnetic flux component and calculating its rate of change (differential value), and is therefore susceptible to the effects of calculation errors, noise, etc. Increasing the averaging time is effective in improving estimation accuracy, but in that case the estimation time becomes longer, making it difficult to restart the motor from idle in a short time.

[0012] In principle, this invention is based on the speed estimation using self-excited vibration of Patent Document 1. An impulse voltage is applied to an induction motor in an idling state, and the resulting current is positively fed back to cause an oscillation phenomenon. At this time, a load voltage command is intentionally set relative to the inverter voltage command value so that the self-excited vibration is maintained stably. As a result, a continuously oscillating current is generated stably, and the rotation speed of the induction motor is estimated from the frequency of this oscillating current.

[0013] In order to solve the above problems, one representative drive control device for an induction motor according to the present invention is a drive control device for an induction motor comprising an inverter for driving the induction motor, a current detector for detecting the primary current of the induction motor as a detected primary current, a self-excited oscillator for generating a voltage command for the inverter and causing the primary current of the induction motor to self-oscillate, and an initial speed estimator for estimating the initial speed of the induction motor based on the self-excited detected primary current, and a voltage compensator for superimposing an additional voltage command on the voltage command, and the initial speed estimator estimates the initial speed under the condition in which the voltage compensator has superimposed the additional voltage command on the voltage command.

[0014] According to the present invention, it is possible to realize a drive system for an induction motor that can instantaneously estimate the rotation speed over a wide range of rotation speeds even when the induction motor is running idle in a de-energized state, and that can smoothly transition to actual operation. Objects, configurations, and effects other than those described above will become apparent from the description of the following embodiments.

[0015] 1 is a block diagram showing the configuration of an induction motor driving device according to a first embodiment; FIG. 2 is a block diagram showing the internal configuration of a self-excited oscillator; FIG. 3 is a diagram showing a general equivalent circuit for one phase of an induction motor; FIG. 4 is a diagram showing impedance characteristics of an induction motor versus rotational frequency; FIG. 5 is a waveform diagram showing a schematic representation of distortion of a voltage waveform due to dead time; FIG. 6 is a diagram showing the principle of correcting an error voltage due to dead time using a waveform; FIG. 7 is a block diagram showing the internal configuration of a voltage corrector; FIG. 8 is a diagram showing waveform graphs from an experiment confirming self-excited oscillation with or without overcompensation of the dead time compensation amount; FIG. 9 is a diagram showing various waveform graphs measured in an actual machine test of initial speed estimation using the first embodiment; FIG. 10 is a block diagram showing the configuration of an induction motor driving device which is a modified example of the first embodiment; FIG. 11 is a diagram showing examples of waveforms of voltage compensation amounts output by a voltage disturbance generator; FIG. 12 is a block diagram showing the configuration of a controller of an induction motor driving device according to a second embodiment; FIG. 13 is a block diagram showing the configuration of a dq coordinate current control unit in a controller of an induction motor driving device according to a third embodiment; FIG. 14 is a block diagram showing the configuration of an initial speed estimator employed in a fourth embodiment; FIG. 15 is a block diagram showing the configuration of a dq coordinate current control unit employed in a fifth embodiment; FIG. 16 is a diagram showing a railway electric vehicle according to a sixth embodiment;

[0016] Hereinafter, with reference to the drawings, a first to sixth embodiments of the present invention will be described. Note that the present invention is not limited to these embodiments. In addition, in the drawings, the same parts are denoted by the same reference numerals.

[0017] Fig. 1 is a block diagram showing the configuration of an induction motor drive device according to a first embodiment of the present invention. The induction motor drive device shown in Fig. 1 is intended to drive a three-phase induction motor 1 (hereinafter, three-phase induction motor will be abbreviated as "induction motor") and is broadly composed of a controller 2, an inverter 3, and a current detector 4. Note that some of the component names shown in Fig. 1 are abbreviated, and the component names with corresponding symbols for each component described below are the official names.

[0018] 1 performs a switching operation based on a PWM (pulse width modulation) signal output from the controller 2, and applies a pulse width modulated voltage to the induction machine 1. The current detector 4 is a current sensor that detects phase currents flowing through the induction machine 1, and detects at least two of the three phase currents.

[0019] Controller 2 is a controller that performs speed sensorless vector control of induction motor 1, and is equipped with both an "actual operation mode" that realizes normal speed sensorless vector control, and an "initial speed estimation mode" that estimates the initial speed of induction motor 1 before actual operation. Switching between these functions will be explained later, but changeover switches 19a, 19b, and 19c are used to switch to the "A" side of these switches in the "initial speed estimation mode," and to the "B" side of these switches in the "actual operation mode."

[0020] The components of the controller 2 and their functions (operating modes) are described below. αβ coordinate converter 5: Converts the currents of the induction machine 1 detected by the current detector 4 (Iu and Iw in FIG. 1) into currents IαFB and IβFB on two fixed coordinate axes (αβ coordinate axes).

[0021] dq coordinate converter 6: converts the current on the αβ axes converted by the αβ coordinate converter 5 into currents IdFB and IqFB on two rotating coordinate axes (dq coordinate axes).

[0022] dq coordinate current control unit 7: A structural block for controlling the current of induction machine 1, consisting of the following four components (enclosed in the dashed line frame in Figure 1). Current command generator 8: Generates current commands Idr and Iqr on the dq axes. Adder-subtractors 9a and 9b: Calculate the error between the current command and the actual detected current. Phase generator 10: Calculates the phase angle θd for dq coordinate transformation. Current controllers (ACR) 11a and 11b: Control the current on each of the d and q axes. ACR is an abbreviation for "Auto Current Regulator."

[0023] Inverse dq converter 12: converts the output voltages on the dq axes calculated by the current controllers 11a and 11b into voltages VαB and VβB on the αβ axes.

[0024] Inverse αβ converter 13: converts the voltage on the αβ axes into three-phase AC voltage commands Vu*, Vv*, and Vw*.

[0025] Adder-subtractors 9c, 9d, 9e: Add correction voltages to the three-phase AC voltage commands Vu*, Vv*, and Vw*, respectively, and output corrected three-phase AC voltage commands Vud*, Vvd*, and Vwd*.

[0026] PWM generator 14: performs pulse width modulation (PWM) based on the corrected three-phase AC voltage commands Vud*, Vvd*, and Vwd* to generate gate signals for inverter 3.

[0027] Voltage corrector 15: Calculates correction voltages Vcu, Vcv, and Vcw for the three-phase AC voltage commands Vu*, Vv*, and Vw*.

[0028] Voltage correction amount setter 16: Sets the voltage correction amount in the initial speed estimation mode. Voltage correction amount setter 17: Sets the voltage correction amount in the actual operation mode.

[0029] Self-excited oscillator 18: Generates voltages VαA and VβB that cause the primary current of induction machine 1 to self-oscillate at the rotation frequency based on the current on the αβ axes.

[0030] Changeover switches 19a and 19b: Switches the voltage applied to the induction machine 1 between the initial speed estimation mode and the actual operation mode. Changeover switch 19c: Switches the amount of correction when performing voltage correction on the induction machine 1 between the initial speed estimation mode and the actual operation mode.

[0031] Initial speed estimator 20: In the initial speed estimation mode, the initial speed ω1est is estimated using the currents IαFB and IβFB.

[0032] 2 is a block diagram showing the internal configuration of the self-excited oscillator 18. The self-excited oscillator 18 is made up of the following components: Trigger generator 21: applies a pulse voltage to induce self-excited oscillation when the initial speed estimation starts.

[0033] Oscillating current command generators 22a and 22b: Generate current commands Iαr and Iβr in the initial speed estimation mode. This function will be described later.

[0034] Adder-subtractors 9f, 9g: Calculate the deviations between the current commands Iαr, Iβr and the actual currents IαFB, IβFB.

[0035] Current controllers 11c and 11d: Execute current control so that the current deviation calculated by the adder-subtractors 9f and 9g becomes zero.

[0036] Next, the basic operation of the initial speed estimation mode in the controller 2 according to the first embodiment will be described. With the induction machine 1 in an idling state, the changeover switches 19a, 19b, and 19c are set to the "A" side. As a result, VαB and VβB applied from the dq coordinate current control unit 7 via the inverse dq converter 12 during actual operation are blocked, and the outputs VαA and VβA of the self-excited oscillator 18 are applied to the PWM generator 14.

[0037] First, the self-excited oscillator 18 applies an impulse voltage by the trigger generator 21, which generates a transient current in the induction machine 1 via the inverter 3. This transient current is detected by the current detector 4 and input to the self-excited oscillator 18 as IαFB and IβFB via the αβ coordinate converter 5. The self-excited oscillator 18 calculates voltages VαA and VβA for controlling this transient current and applies them to the PWM generator 14.

[0038] 2, current commands Iαr and Iβr based on these applied voltages VαA and VβA are generated by oscillating current command generators 22a and 22b. Here, oscillating current command generators 22a and 22b are implemented as a transfer function of a first-order lag element based on winding resistance R1 and primary inductance L1, which are motor constants on the primary side of induction machine 1. With this configuration, oscillations of the same frequency as the rotation speed of induction machine 1 are generated in the current of induction machine 1.

[0039] The principle behind this vibration generation will be explained using Figures 3 and 4. Figure 3 shows a typical equivalent circuit for one phase of an induction motor. Figure 4 shows the impedance characteristics of an induction motor versus rotational frequency, with the horizontal axis representing rotational frequency ω and the vertical axis representing impedance |z|.

[0040] As shown in the equivalent circuit, one phase of an induction motor is composed of primary winding resistance R1, leakage inductance Lσ, excitation inductance Lm, and secondary winding resistance R2 / s. Note that "s" represents "slip," and in this equivalent circuit, the combination of Lσ and Lm forms the primary inductance L1. Here, slip s is determined by the rotational frequency ωr of the induction motor (frequency converted to electrical angular frequency) and the frequency ω1 of the AC voltage applied to the induction motor, and is defined as s = (ω1 - ωr) / ω1.

[0041] As the rotational frequency ωr approaches the frequency ω1 of the applied AC voltage, the slip s approaches zero and the value of R2 / s increases. As a result, the equivalent circuit of the induction motor can be considered a first-order lag element consisting of a series circuit of R1 and L1. Furthermore, when the slip s is "0" (when the rotational frequency ωr and the applied frequency ω1 match), the impedance of the induction motor is at its maximum. This impedance characteristic can be considered a vibration element with a resonance point, as shown in Figure 4.

[0042] 2 generate current commands Iαr and Iβr for inducing oscillation (self-oscillation). These Iαr and Iβr are outputs of a first-order lag element formed by the induction motor constants R1 and L1, which are the resonance conditions. Therefore, by making the actual currents IαFB and IβFB follow these current commands Iαr and Iβr, oscillation of the rotation frequency is induced.

[0043] However, to induce vibration (self-oscillation), a sufficient current needs to flow through the excitation circuit of the induction machine 1, but if this current is insufficient, the current may decay without causing vibration. In particular, when the rotation speed of the induction machine 1 is low or when the electrical time constant of the induction machine 1 is long, self-oscillation is unlikely to occur.

[0044] Countermeasures for this include lengthening the application time of the voltage pulse given by the trigger generator 21, or inserting a high-pass filter to cut out the DC component when detecting the current (Patent Document 1).

[0045] However, since a longer voltage pulse may cause an overcurrent trip or transiently large torque pulsation, it is desirable to set the pulse width as short as possible. Also, if a high-pass filter is introduced, it becomes impossible to estimate the initial speed in the low-speed range, making it difficult to smoothly restart the induction machine 1 at low speeds.

[0046] To solve these problems, the present invention utilizes a voltage compensator 15 shown in FIG. 1. The voltage compensator 15 here is a commonly provided function primarily for correcting error voltages due to the dead time (arm short-circuit prevention period) of the inverter 3. The inverter dead time is a period set to prevent the upper and lower arms of the power semiconductor elements that make up the inverter from being turned on simultaneously, causing an upper and lower short circuit; during this period, the upper and lower power semiconductor elements are turned off simultaneously. However, inserting this dead time causes errors in the output voltage and causes distortion of the current waveform, so correction is necessary.

[0047] Next, the influence of these dead times and the method of correcting them will be described with reference to Figures 5 to 7. Figure 5 is a waveform diagram that schematically shows distortion of the voltage waveform due to the dead time.

[0048] Figure 5(a) shows an ideal U-phase voltage command Vu* calculated by controller 2. When PWM control is performed based on this voltage command Vu* and induction machine 1 is driven, a sinusoidal phase current Iu is generated as shown in Figure 5(b). During dead time, an error voltage depending on the sign of this phase current Iu is generated. However, this dead time is a period during which the upper and lower arms of each phase of the inverter are simultaneously off, so the phase current Iu continues to flow via the freewheel diode. As a result, as shown in Figure 5(c), when the phase current Iu is positive, the voltage Vu decreases, and when the phase current Iu is negative, the voltage Vu increases, resulting in a distorted waveform of the voltage Vu.

[0049] In contrast, Fig. 6 shows the principle of correcting the error voltage due to the dead time using waveforms. Depending on the sign of the phase current Iu shown in Fig. 6(b), a correction voltage Vcu corresponding to the error due to the dead time is added to the original voltage command Vu* (Fig. 6(a)), as shown in Fig. 6(c). That is, if the phase current Iu is positive, "VcB" is added, and if it is negative, "-VcB" is added.

[0050] The waveform of the corrected voltage waveform Vud* is distorted as shown in FIG. 6(d), but by adding a dead time to this waveform, it finally coincides with Vu=Vu*.

[0051] 7 is a block diagram showing the internal configuration of the voltage compensator 15. In the figure, signal comparators 151a, 151b, and 151c are used to compare each phase current (or each phase current command signal) with the set value (=0) of a zero setter 152, and if the phase current is "positive," "1" is output, and if it is "negative," "0" is output.

[0052] The outputs of the signal comparators 151 a, 151 b, and 151 c are used to switch the conditional switches 152 a, 152 b, and 152 c, respectively. If the output of the corresponding signal comparator 151 a, 151 b, or 151 c is “1,” the conditional switches 152 a, 152 b, and 152 c output a set value of “1” set by a value setter 153, and if the output is “0,” the conditional switches 152 a, 152 b, and 152 c output a set value of “−1” set by a value setter 154.

[0053] The outputs (1 or -1) of the conditional switches 152a, 152b, and 152c are multiplied by the value of VcA or VcB set by the voltage correction amount setters 16 and 17 selected by the selector switch 19c in the multipliers 155a, 155b, and 155c to calculate the compensation voltages Vcu, Vcv, and Vcw. That is, in the actual operation mode (the B side of the selector switch 19c), the output is multiplied by the appropriate value VcB for dead time compensation, and in the initial speed estimation mode (the A side of the selector switch 19c), the output is multiplied by a compensation voltage VcB that is greater than the appropriate value VcB in the actual operation mode. Here, the appropriate value VcB is determined based on the length of the dead time, the characteristics of the power semiconductor devices used, the magnitude of the inverter's DC voltage, and the like.

[0054] As described above, the present invention promotes self-oscillation by intentionally adding an additional voltage command to the voltage command sent to the inverter. The dead-time compensation voltage described above is used as an example of a means for generating this additional voltage command. For example, in the initial speed estimation mode, the dead-time compensation voltage value VcA is set to approximately two to four times the appropriate value VcB to intentionally generate an additional voltage command. This additional voltage command, like the pulse voltage output by the trigger generator 21, triggers vibration, enabling it to continuously induce vibration.

[0055] Here, we will show the difference in the promotion effect of self-excited oscillation depending on whether or not there is overcompensation of the dead time compensation amount. Figure 8 shows waveform graphs from an experiment confirming self-excited oscillation depending on whether or not there is overcompensation of the dead time compensation amount. For an induction machine 1 during idling, Figure 8(a) shows the case where the dead time compensation amount is set to the appropriate amount (100%), and Figure 8(b) shows the case where the dead time compensation amount is overcompensated (300%), and the aspects of vibration induction were compared in an actual machine test.

[0056] It can be seen that when the dead time compensation amount shown in FIG. 8( a) is an appropriate amount, no vibration is induced, whereas when the dead time compensation amount shown in FIG. 8( b) is overcompensated and an additional voltage command is intentionally applied, vibration is successfully induced, and an effect of promoting vibration induction is obtained.

[0057] In contrast to the conventional method in which a sufficient excitation current cannot be supplied and vibration cannot be induced by simply applying a trigger voltage initially, the present invention makes it possible to continuously supply excitation current by overcompensating the voltage waveform, thereby inducing smooth excitation vibration. This eliminates the need to insert a high-pass filter as shown in, for example, Patent Document 1, and makes it possible to realize initial speed estimation even in lower speed ranges.

[0058] In the present invention, under conditions where sufficient vibration is generated, the initial speed estimator 20 calculates the period of the oscillating current using at least one of IαFB and IβFB, which are obtained by coordinate-transforming the detected current, and estimates the initial speed ω1est.

[0059] 9A and 9B are graphs showing various waveforms measured in an initial speed estimation test using a real machine in Example 1. Figure 9A shows various waveforms measured in an initial speed estimation test using a real machine in Example 1 in which the idling speed Nr of the induction machine 1 is 300 r / min (Figure 9A), and in Example 9B shows various waveforms measured in an initial speed estimation test using a vibration induction method in Example 1 in which the idling speed Nr of the induction machine 1 is 900 r / min (Figure 9B). As is clear from Figures 9A and 9B, the initial speed estimation method of Example 1, which utilizes the resonance characteristics of the induction machine 1, makes it possible to instantaneously estimate the idling speed of the induction machine 1.

[0060] Next, the operational mode for transitioning from the initial speed estimation mode to the actual operation mode will be described. Once the initial speed estimation is complete, the changeover switches 19a, 19b, and 19c shown in Fig. 1 are switched to "B" to transition to the actual operation mode. During actual operation, the dq coordinate current control unit 7 operates, and current commands Idr and Iqr on the dq axes output by the current command generator 8 are used to provide the excitation current command and torque current command that are originally required for the induction machine 1, and the current controllers 11a and 11b control the actual motor current to follow these commands.

[0061] Furthermore, the phase θd of the dq coordinates required for actual operation is calculated by a phase generator 10 provided in the dq coordinate current control unit 7. The phase generator 10 performs sensorless control to estimate speed and calculate phases using ω1est determined by the initial speed estimator 20 as an initial value. Various methods have been proposed for sensorless speed estimation during actual operation, but as long as the method estimates the rotational speed of the induction machine 1, the initial speed estimated by the present invention can be applied as the initial value.

[0062] Next, a modification of the first embodiment will be described. Fig. 10 is a block diagram showing the configuration of a drive device for an induction motor which is a modification of the first embodiment. In the first embodiment shown in Fig. 1, as a means for applying an additional voltage to promote the vibration induction described above, the dead time compensation amount is overcompensated by the voltage corrector 15 shown in Fig. 7. In contrast, in the modification shown in Fig. 10, a superimposed voltage generator 40 is used to superimpose and add voltage command values ​​Vuh*, Vvh*, and Vwh*.

[0063] 11 is a diagram showing an example of each waveform of the superimposed voltage amount output by the superimposed voltage generator 40. From the top of Fig. 11, the waveforms are those of the superimposed voltage modulation factor command Vch*, the U-phase phase voltage command Vuh*, the V-phase and W-phase phase voltage commands Vvh* and Vwh*, and the UV-phase line voltage Vuvh.

[0064] In this modification, as shown in the figure, a periodically changing modulated wave command is given to each phase based on a PWM carrier signal, and a high frequency voltage is applied to the induction machine 1. As a means for generating an additional voltage to promote vibration induction, not only is the dead time compensation amount described above overcompensated, but the same effect can also be obtained by using a configuration that generates a superimposed voltage as shown in Figures 10 and 11.

[0065] Furthermore, the means for generating the additional voltage is not limited to the above-mentioned method of overcompensating the dead time voltage compensation amount or the modified method of applying a high frequency voltage based on a PWM carrier signal, but other methods such as applying white noise may also be used.

[0066] As described above, according to the first embodiment of the present invention, it is possible to instantaneously estimate the rotation speed of an induction motor that is running idle in a non-energized state over a wide speed range.

[0067] Fig. 12 is a block diagram showing the configuration of a controller 2B of a driving device for an induction motor according to a second embodiment of the present invention. The controller 2B according to the second embodiment shown in Fig. 12 is used in place of the controller 2 according to the first embodiment shown in Fig. 1, and components with the same numbers as those shown in Figs. 1 and 2 have the same functions.

[0068] The main components newly added to the controller 2B according to the second embodiment and their functions (operating modes) are described below. dq coordinate current control unit 7B: Controls the current of the induction machine 1 throughout the initial speed estimation mode and the actual operation mode.

[0069] Arc tangent calculator 23: Calculates the phase θdA of the current commands Iαr and Iβr that induce vibration in the induction machine 1. The arrangement and connections of the other components are also appropriately added or changed.

[0070] In the driving device for the induction motor according to the first embodiment, the self-excited oscillator 18 for estimating the initial speed and the current control unit for the actual operation mode are separately installed and operate independently. However, in order to smoothly transition from the initial speed estimation of the induction motor 1 to the actual operation as a series of operations, it is desirable to share the coordinate conversion unit and the current control unit.

[0071] Furthermore, the self-excited oscillator 18 used in the first embodiment performs current control in the αβ-axis coordinate system. However, the current command in the αβ-axis coordinate system has a sinusoidal waveform, resulting in poor current tracking. In particular, when the initial speed is high, steady-state deviation remains unless the current control response is set to a high gain, resulting in a phenomenon in which no vibration is induced in the worst case. Furthermore, in a large-capacity induction machine, the inverter capacity used is large, making it difficult to increase the switching frequency, making it difficult to achieve a high gain in the current control response.

[0072] Therefore, in the second embodiment, further improvements are made to solve the problems of the first embodiment. Specifically, as shown in Fig. 12, the current controllers 11c and 11b, which are components of the self-excited oscillator 18 (Fig. 2) used in the first embodiment, are eliminated, and current control here is performed by sharing the current controllers 11a and 11b used in the actual operation mode. To achieve this, the current controllers 11a and 11b are switched over to the current commands by using the changeover switches 19a and 19b. In other words, in the initial speed estimation mode, the current controllers 11a and 11b can be said to correspond to the current controllers 11c and 11b of the self-excited oscillator 18 (Fig. 2).

[0073] In the actual operation mode, the changeover switches 19a and 19b are set to the B side, and the outputs Idr and Iqr of the current command generator 8 are used as in the first embodiment. On the other hand, in the initial speed estimation mode, the changeover switches 19a and 19b are set to the A side, and IdrA and IqrA obtained by coordinate-transforming Iαr and Iβr by the dq coordinate converter 6 are used.

[0074] As a result, even when switching from the initial speed estimation mode to the actual operation mode, the outputs of the current controllers 11a and 11b are continuously changed, making it possible to perform an extremely smooth switching.

[0075] Furthermore, the converted coordinates of the dq axes in the initial speed estimation mode are calculated using a newly added arc tangent calculator 23. When vibration is induced, the current commands Iαr and Iβr oscillate with a phase difference of 90 degrees from each other, so if the phase θdA of both commands is calculated using the arc tangent and coordinate conversion is performed using this, the current commands Iαr and Iβr are converted into commands IdrA and IqrA on the rotating coordinate system.

[0076] Since these dq-converted IdrA and IqrA become DC amounts, the deviation can be reduced to zero by the current controllers 11a and 11b. That is, unlike in the first embodiment, there is no need to significantly increase the current control gain to maintain vibration, and vibration can be maintained stably up to high speeds. Furthermore, when switching to the actual operation mode, the changeover switch 19d is used to switch to the phase θd calculated by sensorless control.

[0077] As described above, according to the second embodiment of the present invention, switching from the initial speed estimation mode to the actual operation mode becomes smoother, and stable estimation of the initial speed can be achieved even when the initial speed is high.

[0078] Fig. 13 is a block diagram showing the configuration of a dq coordinate current control unit 7C in a controller of a driving device for an induction motor according to a third embodiment of the present invention. In the third embodiment, this dq coordinate current control unit 7C is used in place of the dq coordinate current control unit 7B of the second embodiment shown in Fig. 12, and the components shown in Fig. 13 that are numbered the same as the components shown in Fig. 12 have the same functions.

[0079] The newly added components of the dq coordinate current control unit 7C according to the third embodiment and their functions (operating modes) are described below. d-axis current lower limiter 24: Sets a limit on the minimum value of the current command IdrA. Zero setter 25: Gives a zero command to the current command IqrA.

[0080] In the previous second embodiment, the current was controlled by the same dq coordinate current control unit in both the initial speed estimation mode and the actual operation mode, thereby improving the continuity between these modes and enabling stable self-oscillation even at high speeds.

[0081] On the other hand, in Example 3, the current command in the initial speed estimation mode is modified in order to further shorten the initial speed estimation time and ensure high-speed speed estimation. In the initial speed estimation mode, the current commands Iαr and Iβr on the αβ axes are converted into DC amounts on the dq axes by the dq coordinate converter 6 using θdA calculated by the arctangent calculator 23. At this time, since the coordinate transformation is performed based on the α axis, the d-axis command IdrA is converted into the magnitude (amplitude) of the oscillating current, and IqrA is converted to zero on average. Therefore, to quickly increase the amplitude, it is advisable to reliably increase the magnitude of IdrA.

[0082] Therefore, by adding a d-axis current lower limiter 24 as in the dq coordinate current control unit 7C shown in Figure 13, the current amplitude is set large from the beginning. This simple addition makes the self-excited vibration occur more quickly and reliably. Furthermore, since the q-axis current command is theoretically zero, a zero setter 25 is used to set it to zero in order to eliminate the influence of unnecessary harmonics and the like.

[0083] As described above, according to the third embodiment of the present invention, the initial speed estimation time is further shortened, and quick startup can be achieved.

[0084] Fig. 14 is a block diagram showing the configuration of an initial speed estimator 20D employed in embodiment 4 of the present invention. By using this initial speed estimator 20D instead of the initial speed estimator 20 in embodiments 1 to 3 shown in Figs. 1, 10 and 12, the driving device for the induction motor according to embodiment 4 is realized.

[0085] To estimate the initial speed, for example, as shown in Patent Document 1 and Non-Patent Document 1, the period of the current waveform that oscillates after self-oscillation is obtained is measured, and the initial speed is calculated from the period. In Examples 1 to 3 of the present invention, the initial speed is estimated using a similar method. However, if the initial speed is extremely slow, measuring the period as described above takes a long time, which may make it difficult to achieve short-time startup.

[0086] To solve this problem, in the fourth embodiment, an initial speed estimator 20D shown in FIG. 14 is introduced. The components of the initial speed estimator 20D according to the fourth embodiment and their functions (operational modes) are described below. Arc tangent calculator 23b: Calculates the current phase angle θi1 between the detected current values ​​IαFB and IβFB based on the two values. Alternatively, the current phase angle θi1 may be calculated based on the current commands Iαr and Iβr generated to cause the primary current to self-oscillate as described above. Adder-subtractor 9i: Calculates the deviation between the current phase angle θ1i and the phase angle (internal phase) θi2 assumed within the initial speed estimator 20D. Here, the initial value of the phase angle (internal phase) θi2 is set to "0."

[0087] PI controller (proportional integral controller) 26: receives the deviation of the adder / subtractor 9i as an input and adjusts the output so that the deviation becomes zero. Integrator 27: integrates the output of the PI controller 26 to calculate the phase angle θi2.

[0088] In this initial speed estimator 20D, the internal phase θi2 is adjusted to match the current phase θi1 by the PI controller 26. As a result, the output of this PI controller 26 matches the estimated value ω1est of the rotational frequency of the induction machine 1.

[0089] Therefore, by using the initial speed estimator 20D according to the fourth embodiment, it is possible to reliably estimate the speed within the convergence time of the set response of the PI controller 26 without measuring the period of the self-oscillating current.

[0090] As described above, according to the fourth embodiment of the present invention, it is possible to estimate the initial speed of the induction machine 1 in a short time over a wide range of rotational speeds.

[0091] Fig. 15 is a block diagram showing the configuration of a dq coordinate current control unit 7E employed in a fifth embodiment of the present invention. By using this dq coordinate current control unit 7E in place of the dq coordinate current control units 7B and 7C shown in Figs. 12 and 13, the driving device for the induction motor according to the fifth embodiment is realized. Here, among the components shown in Fig. 15, those with the same numbers as those shown in Figs. 12 and 13 perform the same functions.

[0092] The components of the phase generator 10E employed in the dq coordinate current control unit 7E according to the fifth embodiment and their functions (operating modes) are described below. Sensorless speed estimator 30: In actual operation mode, this estimates and calculates the drive frequency ωr from the motor constants of the induction machine 1 based on the speed electromotive voltage of the induction machine 1. Adder-subtractor 9j: Adds the slip frequency ωs to the rotational angular frequency ωr calculated by the sensorless speed estimator 30 to calculate the drive angular frequency ω1.

[0093] Slip calculator 31: Calculates the slip ωs of the induction machine 1. Integrator 27: Integrates the drive angular frequency ω1 to calculate the coordinate transformation phase.

[0094] When switching from the initial speed estimation mode to the actual operation mode, the sensorless speed estimator 30 sets the estimated value ω1 obtained by the previous initial speed estimator 20 (or the initial speed estimator 20D according to the fourth embodiment) as the ω1 initial value before transitioning to the actual operation mode.

[0095] Also, in the integrator 27 that calculates the phase, when the mode is shifted to the actual operation mode, the initial phase of the phase angle θd is set to the calculated value of the arctangent calculator 23 as the initial value of θd.

[0096] Therefore, by inheriting the values ​​of the initial speed and initial phase from the initial speed estimation mode, it is possible to smoothly transition to the actual operation mode without causing torque shock or the like.

[0097] As described above, according to the fifth embodiment of the present invention, it is possible to perform initial speed estimation in a short time and to transition to the actual operation mode without any transient phenomenon.

[0098] 16 is a diagram showing a railway electric vehicle according to a sixth embodiment of the present invention, which is an example in which the induction motor drive device according to the first to fifth embodiments is mounted. Here, the same components as those in the first embodiment shown in FIG. 1 are designated by the numbers 2, 3, and 4. However, with regard to controller 2, a controller other than those employed in the second to fifth embodiments may be used.

[0099] The railway electric vehicle according to the sixth embodiment is composed of a car body 100, a controller 2, an inverter 3, a current detector 4, a pantograph 102, and induction machines 1a to 1d mounted on the car body 100, and transmits and receives electric power from an overhead line 101 via the pantograph 102, and runs on a track 103. Note that in railway electric vehicles powered by induction motors, it is common for a single inverter to drive multiple induction machines in parallel, and in Fig. 16 four induction machines 1a to 1d are driven.

[0100] Furthermore, in electric railway vehicles, for example, when the distance between stations on a line is long, it is common to perform "coasting" in which the inverter output is shut off. During this coasting period, the vehicle body moves by inertia, and the induction machines (e.g., induction machines 1a to 1d) enter a non-energized idle state. When accelerating or decelerating again from this state, the induction machines must be smoothly restarted.

[0101] Therefore, by using any of the driving devices for an induction motor according to the first to fifth embodiments of the present invention, it becomes possible to smoothly restart the motor from an idling state, improving the operability of the railway electric vehicle. In addition, since the initial speed can be reliably estimated over a wide rotational speed range, it is highly useful.

[0102] As described above, according to the sixth embodiment of the present invention, it is possible to realize a railway electric vehicle that is capable of smooth restart and has high operability.

[0103] Although the first to sixth embodiments of the present invention have been described above, the present invention is not limited to these first to sixth embodiments, and various modifications are possible within the scope of the present invention.

[0104] 1, 1a, 1b, 1c, 1d... induction machine, 2, 2B... controller, 3... inverter, 4... current detector, 5... αβ coordinate converter, 6... dq coordinate converter, 7, 7B, 7C, 7E... dq coordinate current control section, 8... current command generator, 9a, 9b, 9c, 9d, 9e, 9f, 9g, 9h, 9i... adder / subtractor, 10, 10E... phase generator, 11a, 11b, 11c, 11d... current controller, 12... inverse dq converter, 13... inverse αβ converter, 14... PWM generator, 15... voltage compensator, 16, 17... voltage compensation amount setter, 18... self-excited oscillator, 19a, 19b, 19c, 19d... changeover switch, 20, 20D ...initial speed estimator, 21...trigger generator, 22a, 22b...oscillating current command generator, 23, 23b...arc tangent calculator, 24...d-axis current lower limiter, 25, 152...zero setter, 26...PI controller (proportional integral controller), 27...integrator, 30...sensorless speed estimator, 31...slip calculator, 40...superimposed voltage generator, 100...car body, 101...overhead line, 102...pantograph, 103...track, 151a, 151b, 151c...signal comparator, 152a, 152b, 152c...conditional switch, 153, 154...value setter, 155a, 155b, 155c...multiplier

Claims

1. A drive control device for an induction motor comprising an inverter for driving an induction motor, a current detector for detecting the primary current of the induction motor as a detected primary current, a self-excited oscillator for generating a voltage command for the inverter and causing the primary current of the induction motor to self-oscillate, and an initial speed estimator for estimating an initial speed of the induction motor based on the self-excited detected primary current, wherein the drive control device for an induction motor has a voltage compensator for superimposing an additional voltage command on the voltage command, and the initial speed estimator estimates the initial speed under a state in which the voltage compensator superimposes the additional voltage command on the voltage command.

2. A drive control device for an induction motor as described in claim 1, wherein the self-excited oscillator generates current commands for two orthogonal axes on a fixed coordinate system based on values ​​on the voltage command, and self-oscillates the primary current by a current control function that makes the detected primary current follow the current commands.

3. A drive control device for an induction motor as described in claim 1, wherein the self-excited oscillator generates current commands for two axes that are orthogonal to each other on a fixed coordinate system based on values ​​on the voltage command on the two axes, converts the generated current commands and the detected primary current into values ​​on two axes that are orthogonal to each other on a rotating coordinate system, and self-oscillates the primary current using a current control function that makes the converted detected primary current follow the converted current command.

4. A drive control device for an induction motor as claimed in claim 3, characterized in that the self-excited oscillator calculates a current phase based on the current commands of the two axes, and converts the generated current command and the detected primary current into values ​​of two axes perpendicular to each other on the rotating coordinate system using the calculated current phase.

5. A drive control device for an induction motor according to claim 3 or 4, further comprising a limiter for setting a minimum limit value for the current command converted into values ​​of two axes perpendicular to each other on the rotating coordinate system.

6. A drive control device for an induction motor as claimed in any one of claims 1 to 5, wherein the initial speed estimator has an arithmetic unit which calculates a current phase based on at least one of a current command for self-oscillating the primary current or the detected primary current, and a controller which matches the current phase output by the arithmetic unit with a current phase assumed within the initial speed estimator, and which estimates the initial speed of the induction motor based on the output of the controller.

7. A drive control device for an induction motor as claimed in claim 6, characterized in that the drive control device for an induction motor transitions to drive control during normal operation of the induction motor, using the speed of the induction motor estimated based on the speed electromotive force of the induction motor as an initial value, based on the estimated initial speed of the induction motor.

8. A drive control device for an induction motor as claimed in any one of claims 1 to 7, characterized in that, when estimating the initial speed, the voltage compensator sets a compensation voltage command for compensating for a dead time error voltage contained in the output of the inverter during normal operation of the induction motor to a value greater than the compensation voltage command, and sets this as the additional voltage command.

9. A railway electric vehicle equipped with a drive control device for an induction motor according to any one of claims 1 to 8.

10. A drive control method for an induction motor, which generates a voltage command for an inverter that drives an induction motor, causes a primary current of the induction motor to self-oscillate, and estimates an initial speed of the induction motor based on the self-oscillated primary current, the drive control method for an induction motor comprising the steps of: superimposing an additional voltage command on the voltage command; and estimating the initial speed under a state in which the additional voltage command is superimposed on the voltage command.

11. A drive control method for an induction motor as claimed in claim 10, characterized in that current commands for two axes perpendicular to each other on a fixed coordinate system are generated based on values ​​on the voltage command on the two axes, and the primary current is self-oscillated by a current control function that makes the primary current follow the current commands.

12. A drive control method for an induction motor as claimed in claim 10, comprising the steps of: generating current commands for two orthogonal axes on a fixed coordinate system based on values ​​on the voltage commands on the two orthogonal axes; converting the generated current commands and the primary current into values ​​on two orthogonal axes on a rotating coordinate system; and self-oscillating the primary current by a current control function that makes the converted primary current follow the converted current commands.

13. A drive control method for an induction motor as claimed in claim 12, characterized in that a current phase is calculated based on the current commands of the two axes, and the generated current command and the primary current are converted into values ​​of two axes perpendicular to each other on the rotating coordinate system using the calculated current phase.

14. A method for driving and controlling an induction motor as claimed in claim 12 or 13, characterized in that a minimum limit value is set for the generated current command converted into values ​​of two orthogonal axes on the rotating coordinate system.

15. A drive control method for an induction motor as claimed in any one of claims 10 to 14, comprising the steps of: calculating a current phase based on at least one of a current command for causing self-oscillation of the primary current or the primary current; and estimating an initial speed of the induction motor based on a control mode for matching the calculated current phase with a current phase assumed when estimating the initial speed.

16. A drive control method for an induction motor as claimed in claim 15, characterized in that the estimated initial speed of the induction motor is used as an initial value, and a transition is made to drive control during normal operation of the induction motor, using the speed of the induction motor estimated based on the speed electromotive force of the induction motor.

17. A drive control method for an induction motor as claimed in any one of claims 10 to 16, characterized in that a compensation voltage command for compensating for a dead time error voltage contained in the output of the inverter during normal operation of the induction motor is set to a value greater than the compensation voltage command, and used as the additional voltage command.

Images