JP2013066340A - Motor controller - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a motor controller capable of detecting the current of each phase supplied to a motor with higher accuracy by a single current detection element.SOLUTION: The motor controller drives a motor through a power converter converting DC power into polyphase power, and generates a conduction pattern which follows up the rotor position of a motor. Conduction signal generation means is connected to the DC side of the power converter within the conduction control period, and generates a conduction pattern so that a signal generated in a current detection element corresponds at least to the two-phase current, and current detection means detects a phase current based on a signal generated in the current detection element and the conduction pattern. Current correction means corrects the error included in the current detected, and current control means performs current control for generating a conduction pattern from conduction signal generation means depending on a current command value input and a corrected current. Control switching means outputs a correction value operation command, and switches the period when the current control means performs current control while interlocking with the output state of the command.

Description

  Embodiments described herein relate generally to a motor control device that detects a phase current by a current detection element disposed in a DC portion of an inverter circuit.

  When detecting the current of each phase of U, V, and W in order to control the motor, there is a technique of performing current detection using one shunt resistor inserted in the DC part of the inverter circuit. To detect all three-phase currents using this method, a three-phase PWM signal can be detected so that two or more currents can be detected within one cycle of a PWM (Pulse Width Modulation) carrier. It is necessary to generate a pattern. For example, as shown in FIG. 8 (the carrier is a sawtooth wave), when the U and V phase duties are equal, U + ("+" indicates the upper arm side switching element of the inverter circuit), V + is on, W + When is off, the W-phase current can be detected, but the other phase currents cannot be detected. For this reason, as shown in FIG. 9, it is conceivable that a current of two or more phases can always be detected by shifting the phase of the PWM signal of a certain phase (in this case, the W phase).

Japanese Patent No. 3447366

  However, when the PWM signal of each phase is sequentially shifted for current detection, there is a problem that the detected current detected at a specific timing has an error with respect to the average current (the median value of the ripple) flowing to the motor. This problem will be described by comparing a three-shunt type current detection method and a one-shunt type current detection method in which a PWM signal is shifted. FIG. 10 shows a current detection method using three shunts, and the PWM signal at this time is as shown in FIG. That is, since there is a shunt resistor on the lower arm side of the three-phase bridge (inverter circuit), the PWM signal pattern generated by a general triangular wave comparison method, in FIG. 11, extends in both directions with reference to the intermediate phase of the carrier cycle Motor control can be performed by the PWN pulse.

  At this time, a three-phase motor current waveform in the PWM cycle is shown in FIG. Focusing on the U-phase current, the current basically increases during the high-level period of the U-phase PWM pulse, and decreases during the low-level period. Also, the rate of increase in the period in which only the U-phase PWM pulse is at the high level is greater than the period in which the PWM pulses in the other phases are simultaneously at the high level. This is because the voltage applied to the motor becomes larger in the former period.

  The magnitude relationship of each phase PWM pulse changes according to the duty, but the increase / decrease in each phase current follows the rules described above. When the current detection timing is set to a trough of the carrier waveform that is an intermediate phase of the PWM cycle, the detected current of each phase becomes an average value (median value) of values that change within the PWM cycle.

  Next, a current detection method using one shunt will be described. FIG. 12 shows a circuit configuration, and FIG. 13 shows a PWM pattern. In the 1-shunt current detection method, the current cannot be detected because the motor current does not flow through the shunt resistor inserted in the DC bus in the section where all three phases are H or L. Therefore, a method of shifting the rising phase and the falling phase of each phase PWM pulse so that at least two phases of current can be detected within one cycle is often used.

  At this time, when the current change in each phase is considered in the same manner as in the case of the three shunt system, the result is as shown in FIG. Each phase PWM pulse width and the slope of the current increase when the PWM pulse of a phase shows a high level are the same as in FIG. 11, but all phase pulses are simultaneously high or low by shifting the phase of each phase pulse. The level period decreases, and the phase current waveforms are different from those in FIG. However, since the sum of the applied voltages in each section in the PWM cycle is the same in FIGS. 11 and 13, the average current is the same.

  Then, the current flowing through the shunt resistor according to such a PWM pattern is as shown in (d) in the figure, and the detected current pattern changes depending on the level indicated by each of the three-phase pulses (see FIG. 14). Here, when the current detection timing (sample timing) is set to two points indicated by arrows in the figure, the W-phase current -Iw can be detected at the timing left of the carrier valley and the V-phase current -Iv can be detected at the right timing. . The current detected at these two points is the value at the time when each phase current is marked with ● in (c) in the figure.

  When the detection value obtained at the above detection timing is compared with the average current of each phase shown in the figure, it can be seen that there is a difference between the two. That is, the phase current detected at an arbitrary timing is different from the average current, and in the case of FIG. 14, the W and V phase currents Iw and Iv are detected as currents having a value larger than the average current. As a result, the U-phase current obtained from the calculation based on the fact that the sum of the three-phase currents is zero becomes a value smaller than the average current.

  In the vector control, the average value of current ripples generated in the PWM cycle is used as in the case of the three shunt method. Therefore, when control is performed using the detected current value as shown in FIG. 13, it is included in the detected current. There is a high possibility that torque ripple and noise will occur when the motor is driven due to the influence of errors. FIG. 15 shows a current waveform when the motor is actually driven by PWM control in which phase shift is performed by the single shunt detection method. The phase current has a ripple with a PWM period. When a current having a value far from the average value of the ripple is detected according to the detection timing, the control includes an error.

  Accordingly, a motor control device is provided that can detect the current of each phase supplied to the motor with higher accuracy by a single current detection element.

  According to the embodiment, the motor control device drives the motor via a power converter that converts direct current to multiphase alternating current by performing on / off control of a plurality of bridge-connected switching elements according to a predetermined energization pattern, The energization pattern is generated so as to follow the rotor position of the motor. When the energization signal generating means generates the energization pattern so that the signal generated in the current detection element connected to the DC side of the power converter within the energization control period corresponds to at least two-phase current, the current detection means The phase current of the motor is detected based on the signal generated in the current detection element and the energization pattern.

  The current correction unit corrects an error included in the current detected by the current detection unit, and the current control unit generates an energization signal generation unit according to the input current command value and the current corrected by the current correction unit. Performs current control for generating an energization pattern, and the control switching unit outputs a correction value calculation command to the current correction unit, and the current control unit performs current control in conjunction with the output state of the command. Switch the cycle.

Functional block diagram showing the configuration of the motor control device according to one embodiment Timing chart showing current detection timing The figure which shows the relationship between the motor applied voltage and electric current to the W phase in the PWM pattern shown in FIG. Flow chart of correction process The figure explaining the switching of each period by a control switching part (the 1) The figure explaining the switching of each period by a control switching part (the 2) The figure explaining that correction according to the magnetic pole position is necessary when applied to an embedded magnet type synchronous motor Figure showing the prior art (Part 1) Figure showing the prior art (Part 2) Diagram showing the configuration of the 3-shunt current detection method Diagram showing PWM signal pattern The figure which shows the structure of 1 shunt electric current detection system Diagram showing PWM signal pattern The figure which shows the electric current which flows into the shunt resistance according to the PWM pattern The figure which shows the electric current waveform at the time of actually driving a motor by PWM control which performed the phase shift by 1 shunt detection system

  Hereinafter, an embodiment will be described with reference to FIGS. 1 to 7. FIG. 1 is a functional block diagram showing the configuration of the motor control device. The current control unit (current control means) 1 is proportional to the difference between the input current command values Id_ref and Iq_ref and the d-axis current Id and q-axis current Iq given from the three-phase → dq coordinate conversion unit 13 described later. When voltage command values Vd and Vq are generated by performing integral (PI) or proportional integral derivative (PID) control, they are output to dq → three-phase coordinate conversion unit 3.

  The current command values Id_ref and Iq_ref are generated, for example, by performing PI or PID control in the same manner as described above for the difference between the speed command value ω_ref given from the outside and the rotational speed ω of the motor 2. Note that the d-axis current command value Id_ref is set to zero when the motor 2 is a permanent magnet motor such as a brushless DC motor and performs all-field operation. Further, the current control unit 1 is given a current control cycle command by the correction control switching unit 12.

  In the motor 2, a position sensor 4 such as a Hall IC or a rotary encoder is arranged on a rotor (not shown), and the sensor signal is output as a magnetic pole position θ via a magnetic pole position detector 5 (however, for example, induced voltage) (A position sensorless method in which the magnetic pole position is estimated by detecting) may be used. The dq → three-phase coordinate conversion unit 3 converts the d-axis and q-axis voltage command values Vd and Vq into the three-phase voltage command values Vu, Vv, and Vw based on the magnetic pole position θ, and outputs them to the PWM signal generation unit 6. .

  The PWM signal generator 6 generates a three-phase PWM signal based on the three-phase voltage command values Vu, Vv, and Vw, and each phase switching element U ±, V ±, constituting the inverter circuit (power converter) 7. A gate drive signal is output to W ±. The inverter circuit 7 is supplied with a DC power supply 8 generated by rectifying and smoothing the AC power supply, and the motor 6 is driven by being supplied with the gate drive signal. The DC current detector 9 is a shunt resistor inserted in the DC part of the inverter circuit 7, for example, the negative bus, similarly to FIG. 12, and outputs a DC current detection signal flowing through the inverter circuit 7 to the phase current detection part 10. To do.

  The phase current detection unit (current detection means) 10 receives a current detection timing signal synchronized with the cycle of the PWM carrier from the PWM signal generation unit 6, and detects each phase current based on the detection timing signal (FIG. 14). Although two phases are directly detected, these detections are performed twice at different timings for two consecutive carrier cycles as described later. Then, three-phase detected current values I (u, v, w) 1 and I (u, v, w) 2 are output to the current correction unit (current correction means) 11.

  An inclination calculation command is given to the current correction unit 11 by the correction control switching unit (correction control switching means) 12, and the current value for each phase is determined from the difference between the detected current values I2 and I1 at the timing when the inclination calculation command is given. The current value of each phase is corrected based on the inclination, and the corrected current values Iu ′, Iv ′, and Iw ′ are output to the three-phase → dq coordinate conversion unit 13. The correction control switching unit 12 commands to turn on / off the tilt calculation command for correction and change the execution cycle of the current control unit 11 depending on whether or not the correction performed by the current correction unit 11 is completed. .

  The three-phase → dq coordinate conversion unit 13 converts the three-phase currents Iu ′, Iv ′, Iw ′ into q-axis currents Id, Iq according to the magnetic pole position θ, and outputs them to the current control unit 1. Although the position sensor 4, the inverter circuit 7, and the direct current detector 9 are configured by hardware, the remaining functional blocks are configured by a microcomputer and software executed by the microcomputer.

  Next, the operation of this embodiment will be described with reference to FIGS. As shown in FIG. 2, in this embodiment, the phase of the duty pulse of each phase PWM is shifted so that a two-phase current can be detected within one carrier cycle. There are various methods for shifting the phase as described above. As an example, a carrier having a different waveform for each phase is used in the PWM signal generation unit 6. For the U phase, a sawtooth wave having the maximum amplitude at the start phase (left end in the figure) and the minimum amplitude at the end phase of the cycle is used, and for the V phase, the amplitude is maximum at the center phase of the cycle. Then, a sawtooth wave whose amplitude value decreases is used, and for the W phase, a sawtooth wave whose waveform is opposite to that of the V phase is used. However, in FIG. 2A, for convenience of explanation, a triangular wave having an amplitude value of zero in the center phase of the PWM cycle is shown as a carrier.

  The phase current detected only once by the phase current detector 10 has a current ripple error according to the detection position as described above. In order to correct the error, the current correction unit 11 calculates how much the detected current detection value deviates from the average current value (correction value), and adds the correction value to the detection value to calculate the average current. Ask.

  In order to obtain the above error divergence, it is necessary to detect the current twice. That is, the current gradient and time for a specific energization pattern (eg, U + on, V + on, W + off). For example, in order to correct the W-phase current Iw, the correction value can be calculated if the downward slope of the W-phase current in the period of U + on, V + on, and W + off, and the detection timing and the time until the average current are known.

   Therefore, in order to obtain the current gradient, the current detection cycle is made earlier than the current control execution cycle. For example, current control is executed once per two current detections. As a result, as shown in FIG. 2, since the same PWM pattern (energization pattern) is output twice in continuous carrier cycles (A, B), the timing for detecting current is changed in these two cycles. The W-phase current Iw and the V-phase current Iv are detected at different timings in the first period shown in FIG. 2A and the second period shown in FIG. 2B, and these are defined as Iw1, Iv1, Iw2, and Iv2.

  The current correction unit 11 calculates each difference from these detection values, and the slope of the W-phase current Iw when U + and V + are on and W + is off, and when U + and W + are on and V + is off. The slope of the V-phase current Iv is calculated. FIG. 4 is a flowchart showing the procedure of the correction process. As shown in step S1, the current gradient Id is obtained by dividing the difference between the detection values I2 and I1 by the detection time difference (T2−T1).

Next, the time from the detection timing to the average current value is obtained. Since the PWM pattern to be output is known in advance, a table corresponding to the PWM pattern may be prepared (S2). FIG. 3 shows the relationship between the motor applied voltage and the current to the W phase in the PWM pattern shown in FIG. U, V, and W indicated by solid lines correspond to the three-phase PWM pattern shown in FIG.
The applied voltage (phase voltage) of the W phase is
(1) Maximum in the positive direction when only the W phase is ON (2) 1/2 of (1) when the two phases including the W phase are ON
(3) Maximum in the negative direction when only the W phase is OFF (4) 1/2 of (3) when the two phases including the W phase are OFF
It becomes. The W phase applied voltage shown in FIG. 3 is shown in accordance with the change in the duty, and the W phase applied voltage integrated value is obtained by integrating the W phase applied voltage in accordance with the time. The current divided by the phase inductance is the current.

The phase inductance of the motor 2 substantially corresponds to the current gradient and is obtained by the above-described calculation. The median value of the W-phase applied voltage integrated value in the carrier cycle is a broken line shown in FIG. 3, and can be obtained by averaging the applied voltage integrated values in one cycle. The timing at which the W-phase applied voltage integrated value reaches the applied voltage median is the timing at which the average current flows through the DC current detector 5 (shunt resistor). By obtaining the difference between this timing and the arbitrarily set current detection timing, it is possible to know how much the time from the detection timing to the average current energization timing is different, and this is set as the correction time.
A corrected current value is obtained by calculating a current correction value from the current slope and the correction time for each PWM pattern thus obtained, and adding it to the detected value. That is, as shown in step S3 of FIG. 4, the average value I ′ is obtained by adding a term obtained by multiplying the gradient Id by the correction time to the detected value I1.

  Next, the operation of the correction control switching unit 12 that manages execution of the above-described correction sequence will be described. Since the current control period is made slower than the current detection period in order to detect the current gradient, the current control period actually becomes slower than the limit due to the performance of the hardware. For this reason, the correction current switching unit 12 performs control so as to switch execution / stop of the correction calculation while monitoring the correction state of each phase current.

  Specifically, since the correction value of the phase current is not obtained in the initial stage through operation, a command for the current control cycle (twice the current detection cycle) for obtaining the correction value, and a correction calculation command Are output (see FIG. 5A). Then, when there is no detection error because the correction value is obtained, the correction calculation is stopped, and the current control period and the current detection period are set to the same period (see FIG. 5B). That is, the phase current is detected once for each PWM pattern. And the electric current correction part 11 correct | amends an electric current for every carrier period using the correction value already obtained. By switching in this way, it is possible to perform current control without impairing responsiveness using the corrected current with high accuracy.

  As a condition for switching from FIG. 5A to FIG. 5B, the current gradient of each phase is obtained as shown in FIG. 6, and the correction effect is sufficiently reflected (for example, control start) Switching may be performed after a predetermined time elapses from the point in time, but switching may be performed after obtaining more current samples in order to take into account conditions such as temperature characteristics.

  In this embodiment, for the sake of simplicity, a surface magnet type permanent magnet synchronous motor is assumed in which the phase inductance of the motor 2 does not change depending on the magnetic pole position of the motor 2. However, it can also be applied to an embedded magnet type synchronous motor. In this case, since the current gradient for correction≈change in the inductance due to the magnetic pole position is expected, correction according to the magnetic pole position is required (see FIG. 7). Specifically, the above correction processing is performed for each magnetic pole position and angle. However, if the angular resolution is increased, the necessary data becomes enormous, so it is necessary to determine the angular resolution from the required current detection accuracy. For example, in a set of a motor and inverter circuit, sufficient current detection accuracy (an error of about ± 2 to 5%) can be obtained even if correction is performed with an angular resolution of about π / 6, compared to a case where correction is not performed. Is known experimentally.

  As described above, according to the present embodiment, the PWM electric signal generation unit 6 is connected to the DC side of the inverter circuit 7 within the carrier cycle of PWM control, and the signal generated in the DC current detector 9 (shunt resistor) is The PWM signal energization pattern is generated so as to correspond to the two-phase current, and the phase current detection unit 10 detects the phase current of the motor 2 based on the signal generated in the DC current detector 9 and the energization pattern.

  The current correction unit 11 corrects an error included in the current detected by the phase current detection unit 10, and the current control unit 1 responds to the input current command value and the current corrected by the current correction unit 11. The PWM energization signal unit 6 performs current control for generating an energization pattern, and the control switching unit 12 outputs a correction value calculation command to the current correction unit 11 and interlocks with the output state of the command. The current control unit 1 switches a cycle for performing current control. As a result, even if the detected current does not match the average current during the PWM cycle by adopting the single shunt current detection method, the current is accurately detected by correcting so that the average current is obtained, and the torque ripple is applied to the motor 2. Can be prevented.

  When the control switching unit 12 outputs the correction value calculation command, the control switching unit 12 sets the PWM control cycle so that the PWM signal generation unit 6 outputs the same energization pattern twice in succession, and the phase current detection unit 10 The in-phase current is detected twice at different timings. Thereby, it can correct | amend based on the difference of the electric current value detected twice. Further, the current correction unit 11 can calculate the current gradient from the difference between the current values detected twice, and can perform correction based on the gradient, that is, according to the degree of change in current.

  Furthermore, since the current correction unit 11 obtains the average value of the phase current that changes in the current detection period (carrier cycle) based on the slope of the current, even if the current detected by the one shunt current detection method changes, An average value for the carrier period is obtained, and the motor 2 can be controlled with high accuracy. In addition, when the output of the correction value calculation command is stopped, the control switching unit 12 sets the current control period so that the PWM signal generation unit 6 outputs a different energization pattern each time, and the current correction unit 11 has already been obtained. The current is corrected at each current control period (= carrier period) using the correction value. Therefore, after the correction value is obtained, the drive control of the motor 2 can be performed with high responsiveness.

Although several embodiments of the present invention have been described, these embodiments have been presented by way of example and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.
The method of shifting the PWM duty pulse in the PWM signal generation unit 6 is not limited to the above-described method, and a combination of different waveforms may be used. In addition, for example, after changing the duty command value of each phase using a single carrier such as a triangular wave, the comparison logic of the carrier and the command value is changed in a period different from the period in which the amplitude increases. It may be used.

The same phase current may be detected continuously three times or more.
The DC current detector 9 may be disposed on the positive bus of the inverter circuit 7. Further, the current detection element is not limited to the shunt resistor, and for example, a CT (Current Transformer) may be provided.
As the switching element, an N channel type or P channel type MOSFET, IGBT, power transistor or the like may be used.

  In the drawings, 1 is a current control unit (current control unit), 2 is a motor, 6 is a PWM signal generation unit (energization signal generation unit), 7 is an inverter circuit (power converter), and 9 is a DC current detector (current detection). Element), 10 is a phase current detection unit (current detection unit), 11 is a current correction unit (current correction unit), and 12 is a correction control switching unit (correction control switching unit).

Claims (5)

By switching on and off a plurality of switching elements connected in a bridge according to a predetermined energization pattern, the motor is driven via a power converter that converts direct current into polyphase alternating current, and the rotor position of the motor is followed. In a motor control device that generates an energization pattern,
A current detection element connected to the DC side of the power converter and generating a signal corresponding to a current value;
Energization signal generating means for generating the energization pattern so that a signal generated in the current detection element within an energization control period corresponds to at least two-phase current;
Current detection means for detecting a phase current of the motor based on a signal generated in the current detection element and the energization pattern;
Current correcting means for correcting an error included in the current detected by the current detecting means;
Current control means for performing current control for the energization signal generating means to generate the energization pattern according to the input current command value and the current corrected by the current correction means;
A control switching unit that outputs a correction value calculation command to the current correction unit and switches a cycle (current control cycle) in which the current control unit performs current control in conjunction with an output state of the command. A motor control device. The control switching means sets the current control cycle so that the energization signal generation means continuously outputs the same energization pattern twice or more when the correction value calculation command is output,
The motor control device according to claim 1, wherein the current detection unit detects an in-phase current twice or more at different timings.   3. The motor control apparatus according to claim 2, wherein the current correction means calculates a current gradient from the current value detected twice or more and corrects based on the gradient.   4. The motor control device according to claim 3, wherein the current correction means obtains an average value of phase currents that change in a current detection period based on the slope of the current. When the control switching means stops outputting the correction value calculation command, the current control cycle is set so that the energization signal generating means outputs a different energization pattern each time,
5. The motor control device according to claim 2, wherein the current correction unit performs correction for each current control cycle using a correction value that has already been obtained.

Images