JP2011019399A - Motor control apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress large changes in speed which is generated when a synchronous operation mode is switched to a location feedback operation mode, and to attain uniform acceleration characteristics, regardless of the load torque.SOLUTION: A motor controller is configured by estimating a torque of a permanent magnet motor during the synchronous operation mode for driving the permanent magnet motor and by setting the initial value of a current command value in an operation mode as a location sensorless mode based on information on an estimated value of the torque.

Description

  The present invention relates to a motor control device that stably drives a permanent magnet motor.

To start the permanent magnet motor,
・ As a positioning mode, energize a specific phase to position the rotor. ・ Next, as a synchronous operation mode, synchronize without using information on the rotation angle position of the permanent magnet motor (no position feedback) Drive the motor, gradually increase the output frequency of the inverter, accelerate from the above positioning state to a certain rotation speed,
Next, a method is known in which operation is performed by using the estimated value of the magnetic pole position or information on the rotational angle position by a magnetic pole position sensor or the like as the position feedback operation mode at the rotational speed or higher.

  In this method, when switching the operation mode at a certain rotation speed described above, the virtual rotation position creating the control system in the synchronous operation mode state and the actual rotor position are greatly different or When the continuity of the motor output torque is not maintained before and after, a switching shock is generated in which the rotational speed changes greatly when the rotational speed becomes abnormally high immediately after switching, or conversely becomes abnormally low. The degree of switching shock varies depending on the switching method and the load condition at the time of switching.

  As another shock when the operation mode is switched, there is a problem of a peak current in which the current increases. Regarding a technique for keeping this low, for example, there is a method described in JP-A-2004-222382. In this prior art, as a method for determining the voltage in the synchronous operation mode, the load torque is estimated based on the relationship that the current flowing through the permanent magnet motor decreases as the load torque increases, and the voltage corresponding to the estimated load torque is determined. Apply to permanent magnet motor. After that, when the phase difference between the three-phase phase and the rotation angle position is within the specified phase difference range, the mode is switched to the mode using the rotation angle position information.

JP 2004-222382 A

  In the above prior art, in the synchronous operation mode, the change of the load torque is estimated from the change of the current flowing through the permanent magnet motor in order to determine the voltage. On the other hand, the switching of the operation mode includes the three-phase phase and the rotation angle position. Is a configuration that is performed when the phase difference between and becomes within the specified phase difference range, and does not describe a method of using the load torque estimated in the synchronous operation mode at the time of switching. Also, no countermeasures for large speed fluctuations after switching are shown.

  An object of the present invention is to suppress a large speed change that occurs when switching from the synchronous operation mode to the position feedback operation mode, and to realize uniform acceleration characteristics regardless of the load torque.

One feature of the present invention is that it is activated by a positioning mode in which current is supplied to a specific phase to position the rotor, a synchronous operation mode in which an AC current is not applied to perform position feedback, and an operation mode by position feedback. In the control device for the permanent magnet motor, the ratio of the first current in the current phase energized during positioning and the second current in the phase advanced by 90 degrees in the rotation direction is sequentially changed during the synchronous operation mode. It is characterized by that.

  The other features of the present invention are as described in the claims.

  According to the present invention, it is possible to suppress a large speed change that occurs when switching from the synchronous operation mode to the position feedback operation mode.

1 is an overall configuration diagram of a motor control device according to an embodiment of the present invention. It is a structural example of a power converter circuit. It is an example of a d-axis current controller. It is an example of a q-axis current controller. It is an example of a speed controller. It is an example of the simplified diagram for demonstrating the transition to each operation mode, and the characteristic. It is an example of the test result at the time of a light load when the initial value of ASR is set to zero. It is an example of the test result at the time of heavy load when the initial value of ASR is set to zero. It is an example of the test result at the time of a light load when the initial value of ASR is set to a value corresponding to the acceleration torque and the load torque. It is an example of the test result at the time of heavy load when the initial value of ASR is a value corresponding to the acceleration torque and the load torque. It is a vector diagram for demonstrating axial error (DELTA) (theta) c. It is an example of the simplified diagram for demonstrating the change of each command value in the case of providing the area which makes inverter frequency command value (omega) 1 * constant during synchronous operation mode. It is an example of the block diagram of the electric current detection means in the 2nd Example of this invention. It is an example of the block diagram of the inverter input DC current IDC detection means in the 2nd Example of this invention. It is an example of the wave form diagram for demonstrating the method to reproduce a motor current from a direct current. It is an example of the simplified diagram for demonstrating the change of each command value when changing current phase (theta) p during synchronous operation mode. It is an example of the whole block diagram in the 4th Example of this invention. It is an example of the schematic diagram of a washing machine when this invention is applied to the drive system of a washing machine. It is an example of the schematic diagram of an air conditioner when this invention is applied to the drive system of an air conditioner. It is an example of the block diagram at the time of performing the operation verification of this invention. It is an example of the schematic diagram at the time of applying this invention to an electric oil pump.

  In this embodiment, the position information in the position feedback operation mode is the position sensorless control obtained from the motor voltage command and the motor current information, and the position of the permanent magnet motor rotor in the magnetic flux direction is set to the d axis and then to the rotation direction. With respect to the dq real rotation coordinate system composed of the q axis advanced by 90 degrees, the control dc composed of the control virtual rotor position dc axis and the control position qc axis advanced by 90 degrees in the rotation direction. -Qc control Based on control in a rotating coordinate system. In the following description, the dc-qc coordinate axis is simply referred to as a control axis.

  Further, the description will be made assuming that the permanent magnet motor is non-salient and does not generate reluctance torque.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a basic configuration diagram of a motor control device according to the present invention. The motor control device 1 is broadly divided into the current detection means 12, the d-axis detection current Idc and the q-axis detection current Iqc, which are the outputs thereof, and the calculation is performed and finally applied to the permanent magnet motor (PM) 6. The controller 2 outputs a three-phase voltage command value (Vu *, Vv *, Vw *) to be applied, and applies a voltage according to the three-phase voltage command value (Vu *, Vv *, Vw *) to the permanent magnet motor 6. And a power conversion circuit 5.

  The current detection means 12 includes motor current detection means (7a and 7b) for detecting currents Iu and Iw flowing in the U phase and the W phase in the three-phase AC current flowing in the motor, and the detected motor current is estimated magnetic pole position It is composed of a 3φ / dq converter 8 for obtaining a d-axis detection current Idc and a q-axis detection current Iqc by converting the coordinates from the three-phase axis to the control axis using θdc.

  As shown in FIG. 2, the power conversion circuit 5 includes an inverter 21, a DC voltage source 20, and a driver circuit 23. The inverter 21 is configured by a semiconductor element such as an IGBT (Insulated Gate Bipolar Transistor) or a power MOS FET (Metal Oxide Semiconductor Field Effect Transistor). These semiconductors are used for the U-phase, V-phase, and W-phase upper and lower arms, and the connection points of the upper and lower arms are wired to the permanent magnet motor 6. The inverter 21 performs a switching operation according to the pulsed PWM pulse signals (22a, 22b, 22c) output from the driver circuit 23. By switching the DC voltage source 20, an AC voltage having an arbitrary frequency is applied to the permanent magnet motor 6 to drive the motor.

  The control unit 2 inputs the d-axis detection current Idc, the q-axis detection current Iqc, the d-axis and q-axis voltage command values (Vd * and Vq *), and the actual rotational position (actual rotation) of the rotor of the permanent magnet motor 6. The difference between the axis error Δθc and the axis error command value Δθ * (usually zero) is subtracted from the axis error calculator 10 that calculates the position error (axis error Δθc) between the rotation coordinate axis) and the virtual rotation position (control axis). A PLL controller 13 that adjusts the inverter frequency command value ω1 * so as to be zero, and a control changeover switch (16a and 16b) that switches between a positioning mode, a synchronous operation mode, and a position sensorless mode, which will be described later. In the position sensorless mode, the difference between the frequency command value ω * and the inverter frequency command value ω1 * is obtained by the subtractor 11d, and the q-axis current command value (Iq *) is adjusted so that it becomes zero. A load controller that calculates a q-axis current estimated value Iq ^ using a speed controller 14 including a proportional calculation unit and an integral calculation unit and a d-axis detection current Idc, a q-axis detection current Iqc, and an axis error Δθc in the synchronous operation mode. 15 and the integral term initial value computing unit 17 for computing the integral term initial value I0 of the integral computing unit of the speed controller 14 from the q-axis current estimated value Iq ^ and the d-axis and q-axis current command values (Id * and Iq *) and the difference between the d-axis detection current Idc and the q-axis detection current Iqc are obtained by subtractors 11b and 11c, respectively, and the second current command values (Id ** and Iq ** are set so that these are zero. ), Voltage command value generator 3 for performing vector operation using Id ** and Iq ** and inverter frequency command value ω1 * and outputting Vd * and Vq *, Vd * and Vq * from the control axis Integrating the dq / 3φ converter 4 that outputs the three-phase voltage command values (Vu *, Vv *, Vw *) to be applied to the permanent magnet motor 6 after converting the coordinates to the phase axis, and the inverter frequency command value ω1 * And an integrator 9 for outputting the estimated magnetic pole position θdc.

  Many of the control units 2 are configured by a semiconductor integrated circuit (arithmetic control means) such as a microcomputer or a DSP (Digital Signal Processor).

  Next, each part which comprises the control part 2 is demonstrated.

  The voltage command value generator 3 uses the d-axis and q-axis second current command values (Id ** and Iq **), the inverter frequency command value ω1 *, and the motor constant, as shown in the following equation. Vector calculation is performed and Vd * and Vq * are output.

  Here, in (Expression 1), R is the primary winding resistance value of the permanent magnet motor 6, Ld is the d-axis inductance, Lq is the q-axis inductance, and Ke is the induced voltage constant.

  The axis error calculator 10 calculates the axis error Δθc using the d-axis detection current Idc, the q-axis detection current Iqc, and Vd * and Vq * from the voltage command value generator 3. The axis error Δθc is subtracted from a preset axis error command value Δθ * (usually zero) in the subtractor 11a, and this subtraction value (difference) is proportional-integral controlled by the PLL controller 13 to detect the detection frequency ω1. Is obtained. In a position sensorless mode, which will be described later, this detected frequency ω1 is set as an inverter frequency command value ω1 *, and this is integrated by the integrator 9, whereby the magnetic pole position of the permanent magnet motor 6 can be estimated. The estimated magnetic pole position θdc by this estimation is input to the dq / 3φ converter 4 and the 3φ / dq converter 8, and is used for calculation of each block.

  That is, in the control unit 2 in the present embodiment, the axis error Δθc between the actual rotation coordinate axis of the rotor of the permanent magnet motor 6 and the control axis is calculated, and in other words, the calculated axis error Δθc becomes zero. The inverter frequency command value ω1 * is corrected using a PLL (Phase Locked Loop) method so that the control axis is the same as the actual rotation coordinate axis of the rotor of the permanent magnet motor 6, and the magnetic pole position is estimated.

  Next, the configuration of the current controllers 42 and 43 will be described. FIG. 3 shows the configuration of the d-axis current controller 42. The deviation between the d-axis current command value Id * and the d-axis current detection value Idc given from the host device or the like is obtained by the subtractor 11b, and this is multiplied by the proportional gain Kpd. The output signal of the integration calculation unit 42B that performs integration processing by multiplication is added, and the second d-axis current command value Id ** is output according to the following equation.

  FIG. 4 shows the configuration of the q-axis current controller 43. The difference between the q-axis current command value Iq * and the d-axis current detection value Idc given by the host apparatus or the like by the speed controller 14 is obtained by the subtractor 11c, and this is multiplied by the proportional gain Kpq. The output signal of the integration calculation unit 43B that performs the integration process by multiplying by the integration gain Kiq is added, and the second q-axis current command value Iq ** is output according to the following equation.

  Finally, a block diagram of the speed controller 14 is shown in FIG. When the control changeover switch 16a is on the B side, a proportional operation unit that obtains a deviation between the frequency command value ω * given from the host device etc. and the inverter frequency command value ω1 * by the PLL by the subtractor 11d and multiplies this by the proportional gain Kpa The output signal of 14A and the output signal of the integration calculation unit 14B that performs integration processing by multiplying by the integration gain Kia are added, and a q-axis current command value Iq * is output according to the following equation.

  Here, the integral term initial value I0 of the integral calculation unit when the control changeover switch 16a is switched to the B side is an important control constant in the present invention. Details will be described below.

  The basic operation when starting the permanent magnet motor 6 will be described. FIG. 6 is a simplified diagram showing transition of each operation mode when starting the permanent magnet motor 6. The operation mode includes a positioning mode in which a direct current is gradually passed through a motor winding of an arbitrary phase to fix the rotor of the permanent magnet motor 6 at a certain position, a d-axis current command value Id *, and a q-axis current command value. There are three modes: a synchronous operation mode in which the voltage applied to the permanent magnet motor 6 is determined according to Iq * and the frequency command ω *, and a position sensorless mode in which the inverter frequency command value ω1 * is adjusted so that the axis error Δθc becomes zero. is there.

  In these operation modes, any one of the d-axis current command value Id *, the q-axis current command value Iq *, and the inverter frequency command value ω1 * is changed, or the control changeover switches (16a and 16b) in the control unit 2 are used. Transition to another operation mode by switching. Note that the two control changeover switches (16a and 16b) are simultaneously switched unless otherwise specified.

  In the positioning mode, the control changeover switches (16a and 16b) are set to the A side. That is, the frequency command ω * becomes the inverter frequency command value ω1 * as it is, and the q-axis current command value Iq * 0 given from others such as the host controller becomes Iq * as it is. Inverter frequency command value ω1 * is set to zero to allow direct current to flow through permanent magnet motor 6.

  After the positioning mode ends, the mode changes to the synchronous operation mode. The control changeover switches (16a and 16b) remain on the A side. In the synchronous operation mode, the d-axis current command value Id * is kept at a constant value (this activation method is called Id activation), and the inverter frequency command value ω1 * is increased. As a result, the permanent magnet motor 6 accelerates following the inverter frequency command value ω1 *.

  When the frequency at which position sensorless operation is possible is reached, the control changeover switches (16a and 16b) are set to the B side to shift to the position sensorless mode. Thus, the PLL controller 13 adjusts the frequency command value so that the difference between the shaft error Δθc and the shaft error command value Δθ * (usually zero) becomes zero, and the frequency command value ω * and the inverter frequency command value. The speed controller 14 adjusts the q-axis current command value (Iq *) so that the difference from ω1 * becomes zero. Iq * becomes a value corresponding to the acceleration torque and the load torque, and the permanent magnet motor 6 is accelerated. Thereafter, when the acceleration is finished and the speed becomes constant, the value becomes constant at a value corresponding to the load torque. The d-axis current command value Id * is set to zero during the position sensorless mode because the permanent magnet motor is of a non-salient type.

  6 is actually changed according to the load of the permanent magnet motor 6 and the response frequency of the PLL controller 13, the current controller (42 and 43) and the speed controller 14.

  7 and 8 show changes in the command values, the actual rotation frequency ωr of the permanent magnet motor 6, and the shaft error Δθc when the response frequency of each controller is constant and the load of the permanent magnet motor 6 is changed. . FIG. 7 shows a light load and FIG. 8 shows a heavy load.

  In FIGS. 7 and 8, attention should be paid to changes in the axis error Δc and the rotation frequency ωr. When the load is light (FIG. 7), the shaft error Δθc is substantially zero, and the rotational frequency ωr substantially follows the inverter frequency command value ω1 *. However, at the time of heavy load (FIG. 8), the shaft error Δc is a negative large value from the synchronous operation mode, and the rotational frequency ωr follows the inverter frequency command value ω1 * with a delay. This is because the integral term initial value I0 of the speed controller 14 is zero, so that there is a time delay until the values corresponding to Iq * acceleration torque and load torque are reached. When the shaft error Δc is large, the permanent magnet motor 6 may step out and stop.

  Here, for the purpose of following the inverter frequency command value ω1 * with almost no time delay in the case of a heavy load, values corresponding to the acceleration torque and load torque are set in the integral term initial value I0 of the speed controller 14. The results of this are shown in FIG. 9 and FIG. FIG. 9 shows a light load and FIG. 10 shows a heavy load.

  If attention is paid to changes in the shaft error Δc and the rotational frequency ωr as before, the shaft error Δθc after switching to the position sensorless mode is almost zero in the case of a heavy load (FIG. 10), and the rotational frequency ωr is also It almost follows the inverter frequency command value ω1 *. On the other hand, in the case of a light load (FIG. 9), the axial error Δθc is a positive large value, and an excessive overshoot occurs at the rotational frequency ωr. The amount of overshoot of the rotational frequency ωr is too large, and depending on the application of the motor control device 1, the designed maximum rotational speed may be exceeded, which is a problem.

  7-10, if the integral term initial value I0 of the speed controller 14 is set to an appropriate value according to the load, the rotational frequency ωr follows the inverter frequency command value ω1 * with almost no time delay. However, if the initial value is not appropriate, it can be seen that a time delay occurs. In other words, it is necessary to obtain an appropriate initial value of the speed controller 14 according to the load before the transition to the position sensorless operation.

  Furthermore, it is necessary to make the axis error as close to zero as possible when the position sensorless operation mode is switched. As shown in the above example, in the synchronous operation mode, in the case of Id activation with the d-axis current command value Id * constant, an axis error of 0 degrees with no load and +90 degrees with the maximum load that can be activated occurs. Will do.

  In order to solve these problems, in order to set the integral term initial value I0 of the speed controller 14 to an appropriate value according to the load, it is obtained from the q-axis current estimated value Iq ^ and the position sensorless operation mode switching time point It is an object of the present invention to make the axial error at 0 as close as possible to zero. The integral term initial value I0 may be obtained from a value proportional to the torque of the permanent magnet motor instead of the q-axis current estimated value Iq ^. A value proportional to the torque may be directly input to the speed controller 14, the current controller, or the voltage command value generator 3. This is because the controller 2b can output a control command corresponding to the torque even if a value proportional to the torque of the permanent magnet motor is input to the speed controller 14, current controller, or voltage command value generator 3. Therefore, it is possible to suppress a large speed change that occurs when switching from the synchronous operation mode to the position feedback operation mode, which is the object of the present invention.

  In the present embodiment, reference is made to how to obtain the integral term initial value I0.

  Next, a load estimator that is one of the means for realizing the above object will be described. In this embodiment, the load is the motor output torque. The motor output torque τ and the torque current Iq have the following relationship.

  Here, P is the number of pole pairs, and Ke is an induced voltage constant, both of which are constants. Therefore, in estimating the motor output torque τ, Iq may be estimated. Accordingly, the load estimator 15 shown in FIG. 1 obtains the q-axis current estimated value Iq ^ using the following equation for the d-axis detection current Idc, the q-axis detection current Iqc, and the axis error Δθc.

  If Expression 6 is expressed by a vector diagram, it is as shown in FIG. The first term in (Expression 6) is the magnitude when the current on the qc axis is projected onto the q axis, and the second term is the magnitude when the current on the dc axis is projected onto the q axis. . That is, the q-axis current Iq is obtained using the detected current on the control axis. This is an application of the principle that in the synchronous operation mode, the current phase is automatically shifted according to the load, and Iq corresponding to the load flows.

  In calculating the integral term initial value I0 using the q-axis current estimated value Iq ^ obtained from (Equation 6), there are a plurality of calculation methods shown below, each of which brings about an effect depending on the application. it can.

  [Method 1] The estimated q-axis current value Iq ^ obtained from the values of the d-axis detection current Idc, the q-axis detection current Iqc, and the axis error Δθc at the final point in the synchronous operation mode is used. According to this method, the integral term initial value I0 can be set in accordance with the load state at the time of operation mode switching.

[Method 2] d-axis detection current Idc, q-axis detection current Iqc in the synchronous operation mode, Iq ^ calculated using an average value in a certain section of the axis error Δθc, or each instantaneous instantaneous d-axis detection current Idc, The average value of Iq ^ obtained from the q-axis detection current Iqc and the axis error Δθc is used. According to this method, even when there is load pulsation or the like and the detected value varies, the influence can be minimized by using the average value in the synchronous operation mode.

[Method 3] As shown in FIG. 12, there is provided a section in which the inverter frequency command value ω1 * is constant during the synchronous operation mode, and d-axis detection current Idc, q-axis detection current Iqc, and axis error Δθc in the section. The obtained q-axis current estimated value Iq ^ is used. In this section, since the motor output torque is equal to the load torque with no acceleration torque, Iq ^ corresponding to the load torque can be set as the integral term initial value I0. Furthermore, by setting the section in which the inverter frequency command value ω1 * is constant to one rotation or more of the mechanical angle of the permanent magnet motor 6, it is possible to remove the periodic pulsating torque component that varies for each section.

  With the integral term initial value I0 calculated as described above, the rotational frequency ωr can follow the inverter frequency command value ω1 * without any time delay at any load. As a result, the shock when the operation mode is switched can be greatly reduced. This is because most of the load fluctuates periodically with one rotation of the mechanical angle, but its average value is necessary to reduce the shock at the time of switching.

  However, when torque pulsation occurs in a section shorter than one mechanical angle cycle, the same effect can be obtained by making the frequency command constant for one cycle or more of the pulsating torque.

  A second embodiment of the motor control device 1 according to the present invention will be described with reference to FIGS. The differences from the first embodiment are the configuration of current detection means for obtaining the d-axis and q-axis detection currents flowing through the motor, and how to give a current command value in the synchronous operation mode.

  As shown in FIG. 13, the current detection means 12a includes a current detection circuit 7c and a motor current reproduction calculation that reproduces a three-phase AC current (Iu, Iv, Iw) from the inverter input DC current IDC detected by the current detection circuit 7c. And a 3φ / dq converter 8a that obtains d-axis and q-axis detection currents (Idc and Iqc) by converting from a three-phase axis to a dq axis.

  In this embodiment, the means for detecting the inverter input DC current IDC of the power converter 5a has a configuration using a current detection resistor 45 (FIG. 14). A current detection circuit 46 that detects the inverter input DC current IDC inputs and detects the voltage across the current detection resistor 45 to the operational amplifier 44. The operational amplifier 44 is configured by an IC such as an operational amplifier. When the inverter 21 is configured by a module in which six switching elements such as IPM (Intelligent Power Module) are housed in one package, a shunt resistor is often built in the package for protecting the switching element. In that case, it is not necessary to newly add a current detection resistor for current detection, and the number of parts can be reduced and the space can be saved.

  Next, a motor current reproduction calculator 41a that reproduces a three-phase AC current (Iu, Iv, Iw) from the inverter input DC current IDC detected by the current detection circuit 46 will be described with reference to FIG.

  FIG. 15 shows a reference triangular wave 100, voltage command signals (101a, 101b, 101c) for each phase, PWM pulse signals (22a, 22b, 22c) serving as inverter drive signals for each phase, and input currents (102a for each phase). To d) and the inverter input DC current IDC flowing through the current detection resistor 45. As can be seen from FIG. 15, the inverter input DC current IDC of the power conversion device 5 a changes according to the switching state of the IGBT of each phase. In FIG. 15, the drive signals (22a, 22b, 22c) of each phase IGBT turn on the upper arm of each phase when it is at a high level, and turn on the lower arm of each phase when it is at a low level. Means. Actually, an independent PWM pulse signal is applied to the upper arm and the lower arm of each phase to control the switching operation. Further, in FIG. 15, for the sake of explanation, the dead time is not provided, but actually, the dead time is provided so that the upper and lower arms of each phase are not short-circuited.

  In FIG. 15, in the sections A and D in which the lower arm is turned on only in the W phase and the upper arms of the U phase and the V phase are turned on, the W-phase input current having the reverse polarity can be observed. Further, in the sections B and C in which the lower arms of the V phase and the W phase are turned on and only the U phase is turned on, the U-phase input current having the same polarity can be observed.

  The motor current reproduction computing unit 41a has a sample and hold function, samples and holds the inverter input DC current IDC of the power converter 5a according to the sample and hold signal Tsamp indicating the sections A to D in FIG. By combining the inverter input DC current IDC of 5a, a three-phase AC motor current is output.

  Thus, the inverter input DC current IDC which changes according to the switching state of the IGBT of each phase is observed in the A to D sections, and the inverter input DC current IDC of the power conversion device 5a in each section is combined to obtain 3 A phase AC motor current can be reproduced.

  Next, how to give a current command value in the synchronous operation mode will be described.

  In the positioning mode, as in the first embodiment, a direct current is passed through the d-axis and the permanent magnet motor 6 is fixed at the position of the d-axis. After the positioning is completed, the mode is shifted to the synchronous operation mode. In this embodiment, Id * and Iq * in the synchronous operation mode are obtained from the following equations.

  Here, Ipos is a current value that flows at the final point of the positioning mode, Kpos1 and Kpos2 are current amplitude adjustment gains, ωpos is a change amount of the current phase θp per unit time, and T is a time after transition to the synchronous operation mode. is there.

  In the case of Id activation in which current is supplied only to the d-axis described in the previous embodiment, as already described, the axis error Δθc is 0 ° at no load and + 90 ° at the maximum load. On the other hand, although not described in detail, in the case of Iq activation that flows only to the q axis, when switching to the position sensorless mode, the axis is −90 ° at no load and 0 ° at the maximum load. As a result, an error occurs. As a result, the PLL controller 13 shown in FIG. 1 operates so as to make its axis error zero, and the detection frequency ω1 greatly fluctuates, and correct speed detection cannot be performed. Therefore, the switching shock becomes large.

  In this embodiment, by changing the current phase θp during the synchronous operation mode, the current that has flowed only in the d-axis is reduced so that it also flows in the q-axis. Thus, by changing the current phase during the synchronous operation mode, the maximum value of the axis error Δθc can be reduced, and the switching shock can be reduced.

  Here, depending on the method of determining the current amplitude adjustment gains Kpos1 and Kpos2, there are a plurality of calculation methods shown below, and each can bring about an effect corresponding to the application.

  [Method 1] A method in which Kpos1 = Kpos2 = 1 and the current phase θp = 45 ° when the position sensorless mode is switched. In this case, at position sensorless switching, Id * = Iq *. The state under the maximum load condition is shown in FIG. During the synchronous operation mode, the current phase θp changes from 0 ° to 45 °, and when the mode is switched, the axis error Δθc becomes + 45 °. On the other hand, when no load is applied, the angle is −45 °, and the maximum load range is 45 °.

  [Method 2] A method in which Kpos1 = 1, Kpos2 = Iq ^ / Ipos, and the current phase θp = 90 ° when the position sensorless mode is switched. This method is characterized in that the output of the load estimator shown in FIG. 1, that is, Iq ^ obtained according to (Equation 6) is Kpos2. When the position sensorless mode is switched, since the current phase θp = 90 °, at that time, Id * = 0 and Iq * = Iq ^, and the axial error is almost zero.

  [Method 3] A method in which Kpos1 = Kpos2 = Iq ^ / Ipos and the current phase θp = 90 ° when the position sensorless mode is switched. This method is characterized in that the output of the load estimator shown in FIG. 1, that is, Iq ^ obtained according to (Equation 6) is Kpos1 and Kpos2. The difference from method 2 is that the axis error is almost zero even during the synchronous operation mode.

  [Method 4] The initial stage of the synchronous operation mode is the above-described method 1, and at least the end stage is the above-described method 2 or method 3. This method is an optimal method when the estimation error is included in the axis error Δθc at the initial stage of the synchronous operation mode.

  A third embodiment of the motor control device according to the present invention will be described below.

  An overall configuration diagram of the motor control device 1b in the present embodiment is shown in FIG. The configuration of the load estimator is different from the two embodiments shown above. The load estimator 15a receives the d-axis and q-axis voltage command values (Vd * and Vq *) and the d-axis and q-axis detection currents (Idc and Iqc), and the effective power of the permanent magnet motor by the following equation: And a torque estimated value (τ ^) are calculated.

  Here, in (Equation 8), Wp is the active power, Wcu is the copper loss, ωr is the actual rotational frequency of the permanent magnet motor, R is the primary winding resistance value of the permanent magnet motor 6, and ω * is the frequency command. , P is the number of pole pairs of the permanent magnet motor.

  Further, a q-axis current estimated value (Iq ^) is estimated from τ ^ calculated using (Expression 8) by the following expression.

  Here, in (Equation 9), τ is an estimated torque, P is the number of pole pairs, and Ke is an induced voltage constant.

Using Iq ^ estimated based on (Equation 8) and (Equation 9), as shown in the previous two embodiments,
Calculating the integral term initial value I0 of the integral calculation unit 14B constituting the speed controller 14 when the position sensorless mode is switched;
In the synchronous operation mode, it can be used to create Id * and Iq * by applying to Kpos1 and Kpos2 shown in (Expression 7).

  The present embodiment is based on the idea that the motor output is obtained by subtracting the copper loss from the active power, and the estimated torque is obtained based on this. Since the value used in (Equation 8) does not depend on the rotor position, it can be obtained even when the axis error Δc occurs, and it can be obtained by a simple calculation using only four arithmetic calculations. is there. Further, there is a feature that the torque can be estimated more accurately as the active power is larger than the copper loss.

  A motor control apparatus according to a fourth embodiment of the present invention will be described with reference to FIG. FIG. 18 is a schematic diagram when the motor control device 201 according to the present invention is applied to a drive system of a washing machine. The washing machine 200 includes a washing tub 206 and a stirring blade (pulsator) 205 in a water receiving tub 208, and the washing tub 206 and the stirring blade 205 are driven by a driving motor 203. Which of the washing tub 206 and the stirring blade 205 is driven is switched by the clutch unit 204 during the washing process. Note that the clutch unit 204 may be configured with or without a speed reduction mechanism. The motor control device 201 is driven by applying an AC voltage to the driving motor 203 via the motor wiring 202.

  The washing process of a washing machine can be broadly divided into “washing”, “rinsing”, “dehydration” and “drying”. Each of these processes is characterized in that the frequency command ω * and the start-up time change, and further, the load torque and the moment of inertia change greatly depending on the amount of laundry and the cloth quality. In particular, at the time of starting “washing”, the stirring blade 205 is driven by the driving motor 203 while the laundry is immersed in water, and therefore the load torque seen from the driving motor 203 changes every moment. It is no exaggeration to say that the washing machine has no steady state. In such an application, it is very difficult to uniquely determine the response frequency of the speed controller 14 in particular, and there is a trade-off that if the response frequency is adjusted to some load, the performance deteriorates at other loads. appear. However, by using the load estimator 15 of the present invention, the startup characteristics can be made substantially constant. Further, the d-axis and q-axis detection currents are detected from the inverter input DC current IDC of the power conversion circuit 5a and used for each control, and the inverter frequency command value ω1 * is PLL-controlled so that the calculated axis error Δθc becomes zero. By correcting using the method and estimating the estimated magnetic pole position θdc, the motor current detecting means (7a and 7b) and the position sensor can be omitted, and the washing tub can be made larger.

  A fifth embodiment of the motor control apparatus according to the present invention will be described with reference to FIG. FIG. 19 is a schematic diagram when the motor control device 301 according to the present invention is applied to an air conditioner 300. The air conditioner has an indoor unit 302 and an outdoor unit 303. The indoor unit and the outdoor unit are connected by a pipe 304, and a refrigerant flows through the pipe. The indoor unit includes a heat exchanger 305 and a blower 306, and the outdoor unit includes a heat exchanger 307, a compressor 308, a compressor driving motor 309, and a motor control device 301. . In the air conditioner, the refrigerant flows between the indoor unit and the outdoor unit, and cool air or hot air is sent into the room by the heat exchanger of the indoor unit.

  In such a configuration, the compressor has torque pulsation caused by one rotation of the mechanical angle or due to load characteristics. In addition, there are pressure equalization states where there is almost no pressure difference between the input and output sides of the compressor and differential pressure states where there is a pressure difference. The load torque seen from the compressor drive motor is light in the pressure equalization state and heavy in the pressure difference state. It has the feature of becoming. In such an application, for example, if the motor is started in accordance with the pressure equalization state, a shock at the time of switching the position sensorless mode occurs in the differential pressure state, and the start-up performance deteriorates. Conversely, the motor is started in accordance with the differential pressure state. Then, the start-up performance in the pressure equalization state becomes worse.

  Therefore, by using the load estimator 15 of the present invention, it is possible to estimate the torque commensurate with the load, and smooth start-up can be realized in any pressure state. As a result, “rapid cooling” (or “rapid heating”), which is an important function as an air conditioner, can be realized under any conditions.

  The control unit 2 of the motor control device 1 according to the present invention is often configured by software using a semiconductor integrated circuit such as a microcomputer or a DSP. Therefore, there is a drawback that it is difficult to verify whether the control unit 2 is correctly configured. Therefore, in this embodiment, a method for verifying whether the configuration related to the present invention is operating correctly will be described with reference to FIG.

  The values that need to be measured are the three-phase voltage values (Vu, Vv, Vw), the three-phase current values (Iu, Iv, Iw) output from the motor control device 1, and the magnetic pole position θd of the permanent magnet motor 6.

  The three-phase voltage value can be measured by measuring the voltage between the N side of the DC voltage source 20 and each phase terminal (30a, 30b, 30c). Alternatively, the line voltage of each phase may be measured and calculated therefrom.

  The three-phase current value can be measured by, for example, CT.

  The magnetic pole position θd of the permanent magnet motor 6 can be measured, for example, by attaching a magnetic pole position sensor 52 using an encoder or the like. θd is a position in the direction of the magnetic flux of the rotor of the permanent magnet motor 6, and the direction of the winding magnetic flux generated when a current is passed through the U phase on the stator side is selected as 0 °. The three-phase voltage value and the three-phase current value are substituted into the 3φ / dq converters (8a and 8b), respectively, and the dq-axis voltage value (Vd and Vq) and the dq-axis current value (Id and Iq are used using the magnetic pole position θd. )

  Since the controller 2 does not perform position control in the synchronous operation mode, an axis error between the actual rotation coordinate axis and the control axis occurs. Therefore, the dq axis current changes depending on the load. Therefore, paying attention to the dq-axis voltage values (Vd and Vq), it is confirmed whether the relationship of (Equation 1) is satisfied. For example, when Iq * is set to zero during the synchronous operation mode as in the first embodiment, a voltage corresponding to R × Id * should be output to Vd.

  In the position sensorless mode, attention is particularly paid to the movement of each value when the mode changes. If the load estimator 15 is operating normally, each value changes when the mode changes. Depending on the load, a discontinuous current waveform is observed. Next, only the load of the permanent magnet motor 6 is changed without changing the operating conditions of the positioning mode and the synchronous operation mode, and changes in values when the mode transitions are observed. If Vq at the time of transition to the position sensorless mode increases as the load increases, it can be confirmed that the load estimation is performed normally. Further, the magnetic pole position θd is put into the differentiator 51 to determine the actual rotational frequency ωr of the permanent magnet motor 6. If the rotational frequency ωr follows the inverter frequency command value ω1 * with almost no time delay regardless of the load, the final effect of the load estimation can be confirmed.

  A motor control apparatus according to a seventh embodiment of the present invention will be described with reference to FIG.

  FIG. 21 is an example of a schematic diagram when the motor control device 401 according to the present invention is applied to the electric oil pump 400.

  The electric oil pump 400 adjusts the discharge pressure (pressure) of the hydraulic circuit 402. When the “synchronous operation mode” shown in FIG. 2 is shifted to the “position sensorless mode”, the load estimator 15 is not present. There is a problem that the motor 6 suddenly accelerates and decelerates, and accordingly the oil pressure cannot be kept constant (or a long time is required until the oil pressure becomes constant).

  Therefore, by using the load estimator 15 of the present invention, it is possible to estimate a torque commensurate with the load, and it is possible to quickly keep the hydraulic pressure constant.

  DESCRIPTION OF SYMBOLS 1,1b ... Motor control apparatus, 2, 2b ... Control part, 3 ... Voltage command value preparation device, 5 ... Power conversion circuit, 6 ... Permanent magnet motor, 7a, 7b ... Motor current detection means, 7c ... Current detection circuit, 8, 8a, 8b ... 3φ / dq converter, 9 ... integrator, 10 ... axis error calculator, 11a, 11b, 11c, 11d ... subtractor, 12, 12a ... current detection means, 13 ... PLL controller, 14 ... speed controller, 14A ... proportional calculation unit, 14B ... integral calculation unit, 15, 15a ... load estimator, 16a, 16b ... control changeover switch, 17 ... integral term initial value calculation unit, 20 ... DC voltage source, 21 ... Inverter, 22a, 22b, 22c ... PWM pulse signal, 23 ... driver circuit, 41 ... motor current reproduction calculator, 42 ... d-axis current controller, 43 ... q-axis current controller, 44 ... operational amplifier, 45 ... current detection Resistance, 46 ... electric Current detection circuit 50 ... Verification device 52 ... Magnetic pole position sensor 200 ... Washing machine 203 ... Drive motor, Idc ... d-axis current, Iqc ... q-axis current, Id * ... d-axis current command value, Iq * ... q-axis current command value, Iq ^ ... q-axis current estimated value, I0 ... integral term initial value, Vd * ... d-axis voltage command value, Vq * ... q-axis voltage command value, ω * ... frequency command value, ω1 * ... Inverter frequency command value, ω1 ... detection frequency, ωr ... rotational frequency, Δθc ... axis error, θdc ... estimated magnetic pole position, θp ... current phase, IDC ... inverter input DC current.

Claims (4)

Positioning mode in which current is passed through a specific phase to position the rotor,
Synchronous operation mode in which alternating current is passed and position feedback is not performed,
In the control device of the permanent magnet motor that is activated by the operation mode by position feedback,
A motor control device characterized by sequentially changing a ratio of a first current in a current phase energized during positioning and a second current in a phase advanced by 90 degrees in a rotation direction during a period of a synchronous operation mode. In claim 1,
The motor control apparatus according to claim 1, wherein the first current is a d-axis current, the second current is a q-axis current, the first current is a q-axis current, and the second current is a d-axis current. . In claim 1,
Immediately before switching from the synchronous operation mode to the operation mode by position feedback, the first current and the second current are substantially equal. In claim 1,
In the synchronous operation mode, an estimation calculation of a value proportional to the torque of the permanent magnet motor is performed,
Based on a value proportional to the torque of the permanent magnet motor, the ratio of the first current and the second current immediately before switching from the synchronous operation mode to the operation mode by position feedback is increased as the torque increases. A motor control device characterized by changing to a direction in which the current decreases or a direction in which the second current increases.

Images