JP2016226205A - Control device for electric motor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a control device for an electric motor, which is capable of suppressing occurrence of lock or power swing in the electric motor.SOLUTION: An EOPECU (Electric Operated Pump Equipment Controlling Unit) controls a controlled variable of the electric motor by switching electrification phases in such a manner that two phases in motor coils of three phases of the electric motor become electrification phases and one remaining phase becomes a non-electrification phase. The EOPECU comprises a switching control part that executes magnetic pole switching control for switching the electrification phases based on detection of magnetic pole positions. When electrification states of the motor coils of three phases are switched while a driving motor is under magnetic pole switching control, the switching control part acquires an induction voltage that appears in the non-electrification phase before the detection of the magnetic pole positions (S32), and discriminates based on the acquired induction voltage whether spike noise is appearing (S33 and S34). While it is discriminated that the spike noise is appearing, the switching control part sets a non-detection period in which the magnetic pole positions are not detected, by invalidating detection results of magnetic pole position detection signals S1-S3 (S35).SELECTED DRAWING: Figure 5

Description

  The present invention relates to an electric motor control device.

  When the motor is controlled without a sensor, the energization state of each phase of the motor coil is switched so that two of the three-phase motor coils are energized and the remaining one is non-energized. In sensorless control, there is one in which a magnetic pole position is detected using an induced voltage generated in a motor coil in a non-energized phase, and the energized state is switched on condition that the magnetic pole position is detected. In this case, it is known that the induced voltage generated in the non-energized phase generates edge noise, so-called spike noise, when the motor coil is switched from energization to non-energization. Spike noise is a cause of erroneous detection of the magnetic pole position.

  As a countermeasure against erroneous detection of the magnetic pole position related to such spike noise, a countermeasure described in Patent Document 1 is known. Patent Document 1 discloses determining whether or not spike noise is based on regularity that generates spike noise, and determining whether or not the magnetic pole position can be correctly detected based on the determination result. . And in patent document 1, when it can judge that it is spike noise, the detection of the magnetic pole position regarding spike noise is disregarded and this is not employ | adopted.

JP 2012-80690 A (particularly paragraph [0039])

  By the way, in a motor, spike noise may be generated by deviating from the above regularity due to fluctuations in a load connected to the motor or a power supply voltage as a power supply source of the motor. That is, as in Patent Document 1, when it is determined whether or not the spike noise is based on the regularity, the spike noise that deviates from the regularity cannot be dealt with. Then, the magnetic pole position is erroneously detected due to spike noise generated outside the regularity, and in the worst case, the motor may be locked or stepped out.

  The present invention has been made in view of such circumstances, and an object thereof is to provide an electric motor control device capable of suppressing the occurrence of motor locking and step-out.

  An electric motor control device that solves the above-described problem is a three-phase motor coil of an electric motor having a stator having a three-phase motor coil and a rotor that forms a field, with two phases being energized and the remaining one phase. The control amount of the electric motor is controlled by switching the energization state of the motor coil of each phase so as to be in a non-energized phase. The control apparatus for the electric motor includes a magnetic pole position detector that detects the magnetic pole position of the rotor based on an induced voltage that appears in the non-energized phase when the energized phase is energized, and the magnetic pole position detector that determines the energized state of the three-phase motor coil. A switching control unit that performs magnetic pole switching control that switches based on detection of the detected magnetic pole position of the rotor, and when the switching control unit performs magnetic pole switching control, the magnetic pole position prior to detection of the magnetic pole position by the magnetic pole position detection unit An invalid period setting unit that sets an invalid period for invalidating a detection result of the detection unit for a predetermined period.

  The spike noise described in the above [Background Art] is in the non-energized phase that has been switched to non-energized, and is opposite to the energized state so far (negative direction with respect to positive direction, positive direction with respect to negative direction). It occurs as a current flowing through In this case, the current flowing on the opposite side flows larger than usual, so that the width of the voltage fluctuation also appears larger than the power supply voltage that is the power supply source of the electric motor. That is, the spike noise is characterized in that a voltage larger than the power supply voltage or a voltage smaller than the reference potential is instantaneously generated in the non-energized phase switched from energized to non-energized. The above-described configuration utilizes such features of spike noise.

  That is, according to the above configuration, when the energization state of the three-phase motor coil is switched, the induced voltage generated in the non-energized phase, that is, the phenomenon of spike noise generated in the non-energized phase can be detected. . Therefore, the invalid period can be set based on whether or not spike noise occurs.

  For example, when the period during which spike noise occurs changes for a short time, the invalid period changes for a short period according to the change of such a period. As a result, it is possible to cope with spike noise that appears without restriction due to the length of the generated period. Therefore, erroneous detection of the magnetic pole position related to spike noise can be suppressed, and the occurrence of lock and step-out of the electric motor can be suppressed.

  Specifically, the invalid period setting unit may set the invalid period while the induced voltage appearing in the non-energized phase is larger than the power supply voltage or while the induced voltage appearing in the non-energized phase is smaller than the reference potential. preferable.

  That is, according to the above configuration, the effect can be exerted only while the feature of spike noise appears. Therefore, the period for setting the invalid period can be optimized.

  As described above, spike noise may cause a voltage higher than the power supply voltage or a voltage lower than the reference potential. In other words, when switching from negative energization to non-energization, the spike noise generates a voltage larger than the power supply voltage. In this case, the invalid period can be optimized by setting the period in which the induced voltage appearing in the non-energized phase is larger than the power supply voltage as the invalid period. In addition, when switching from energization in the positive direction to deenergization, the spike noise generates a voltage smaller than the reference potential. In this case, the invalid period can be optimized by setting the period in which the induced voltage appearing in the non-energized phase is smaller than the reference potential as the invalid period.

  In other words, when further optimizing the control device for the electric motor, the switching control unit, for the energized state of the three-phase motor coil, appears in the positive direction energization, the negative direction energization, and the non-energization one by one. The invalid period setting unit sets a period in which the induced voltage appearing in the non-energized phase is larger than the power supply voltage as the invalid period for the non-energized phase after switching from negative energization to non-energization. For the non-energized phase after switching from the energization in the positive direction to the non-energized phase, it is preferable to set a period in which the induced voltage appearing in the non-energized phase is smaller than the reference potential as the invalid period.

  According to the present invention, it is possible to prevent the electric motor from being locked or out of step.

The block diagram which shows schematic structure of the system which mounts EOPECU. The block diagram which shows the structure about EOPECU. The block diagram which shows the structure of the predriver about EOPECU. The chart figure which shows the induced voltage which appears in the non-energized phase of the electric motor at the time of high rotation. The flowchart which shows the process in connection with the magnetic pole switching control which the microcomputer performs about EOPECU. (A)-(c) is a figure which shows the aspect in which a non-detection period is set with respect to the spike noise which arises in various aspects.

  Hereinafter, as an embodiment of an electric motor control device, an electric pump control device that controls an electric motor built in an electric pump mounted on a vehicle will be described as an example. First, an outline of the electric pump will be described.

  As shown in FIG. 1, the electric pump 10 includes an electric motor 10a. The electric motor 10a is a surface magnet type brushless DC motor. The electric motor 10a includes a rotor 10b that rotates about its central axis m, and a stator 10c that is disposed on the outer periphery of the rotor 10b. A permanent magnet is fixed to the surface of the rotor 10b. The permanent magnets are alternately arranged with different polarities (N pole, S pole) in the circumferential direction of the rotor 10b. Such a permanent magnet forms a magnetic field, that is, a field, when the electric motor 10a rotates. In the stator 10c, the phases 10u, 10v, 10w of the three-phase motor coils are arranged in an annular shape.

  The electric pump 10 supplies oil to a CVT (continuously variable transmission) 11 by a built-in electric motor 10a. The CVT 11 is a transmission that changes the rotation of the internal combustion engine 12 (E / G) and transmits it to the wheels. The CVT 11 makes the gear ratio continuously variable (steplessly variable) by the oil pressure. Oil supplied to the CVT 11 is stored in the tank 13. The oil in the tank 13 is pumped up by the drive of the electric motor 10 a built in the electric pump 10 and discharged to the CVT 11. The oil in the tank 13 is pumped up by an oil pump 14 driven by the internal combustion engine 12 and discharged to the CVT 11. The oil in the tank 13 is pumped up by either the electric pump 10 or the oil pump 14 and discharged to the CVT 11. A check valve (not shown) is provided between the electric pump 10 and the CVT 11 and the oil pump 14. The check valve regulates the inflow of oil from the CVT 11 and the oil pump 14 to the electric pump 10 side.

  The internal combustion engine 12 is connected to an EGECU 15 that controls the drive of the internal combustion engine 12. The IGECU 15 is connected to an ISM ECU 16 that instructs automatic stop (or return) of the internal combustion engine 12. The ISM ECU 16 controls a so-called idling stop that automatically stops the internal combustion engine 12 when the vehicle is temporarily stopped.

  The electric pump 10 is connected to an electric pump controller (hereinafter referred to as “EOPECU”) 20 that controls the driving of the electric motor 10a. A battery 17 serving as a power source for the electric motor 10a is connected to the EOPECU 20. The battery 17 is a DC power source that stores electric power generated by an alternator (not shown) driven by the internal combustion engine 12. The battery 17 is also connected to the EGECU 15 and the ISEC ECU 16. An ISECU 16 is connected to the EOP ECU 20. The ISM ECU 16 instructs the EOP ECU 20 to drive or stop the electric pump 10 in accordance with a state such as automatic stop of the internal combustion engine 12.

Next, the electrical configuration of the EOPECU 20 will be described.
As shown in FIGS. 1 and 2, the electric pump control signal Seop is input from the ISM ECU 16 to the EOPECU 20. On the other hand, the EOPECU 20 outputs pump state information Pst indicating the driving state of the electric pump 10 to the ISM ECU 16.

  Specifically, the ISM ECU 16 generates an internal combustion engine control signal Seg that instructs the internal combustion engine 12 to automatically stop or return from the automatic stop, and outputs the internal combustion engine control signal Seg to the EGECU 15. When the temporary stop of the vehicle is detected, the ISM ECU 16 instructs the automatic stop of the internal combustion engine 12 according to the state of the vehicle such as the operation state of the internal combustion engine 12 or the power storage state of the battery 17. Further, when the internal combustion engine 12 is automatically stopped, the ISM ECU 16 instructs the internal combustion engine 12 to return from the automatic stop according to the operating state of the internal combustion engine 12, the state of power storage of the battery 17, and the like. Note that the operating state of the internal combustion engine 12, the power storage state of the battery 17, and the like are determined based on various sensors (not shown) mounted on the vehicle and engine state information Dst acquired from the EGECU 15.

  Further, when outputting the internal combustion engine control signal Seg to the EGECU 15, the ISM ECU 16 generates an electric pump control signal Seop that instructs to drive or stop the electric pump 10 and outputs it to the EOP ECU 20. When instructing the automatic stop of the internal combustion engine 12, the ISM ECU 16 instructs to drive the electric pump 10. In this case, the oil pump 14 is stopped by the automatic stop of the internal combustion engine 12. On the other hand, the ISM ECU 16 instructs to drive the electric pump 10. That is, the ISM ECU 16 instructs the electric pump 10 to be driven so that oil can be supplied to the CVT 11 even after the oil pump 14 is stopped by the automatic stop of the internal combustion engine 12. Further, when instructing the return from the automatic stop of the internal combustion engine 12, the ISM ECU 16 instructs the stop of the electric pump 10. In this case, the oil pump 14 is driven by returning the internal combustion engine 12 from automatic stop. On the other hand, the ISM ECU 16 instructs to stop the electric pump 10. That is, the ISM ECU 16 instructs the electric pump 10 to stop because the oil pump 14 is driven by the return from the automatic stop of the internal combustion engine 12.

  Returning to the description of FIG. 2, the EOPECU 20 includes an inverter 21 that is driven to supply AC power from the battery 17 to the electric motor 10 a. The inverter 21 includes a plurality of switching elements (for example, field effect transistors) 21a to 21f classified as transistors. Specifically, the inverter 21 includes a series unit of three pairs of upper and lower switching elements ((21a, 21b), (21c, 21d), (21e, 21f)) as a basic unit (switching arm). It is configured as a known three-phase inverter connected in parallel. The upstream side of the pair of switching elements is connected to the battery 17, and the downstream side of the pair of switching elements is connected to a reference potential point (grounded). Each connection point of the pair of switching elements is connected to each phase 10u, 10v, 10w of the stator 10c of the electric motor 10a. A diode is connected in parallel to each of the switching elements 21a to 21f. The inverter 21 is provided with a capacitor (not shown) that smoothes the power supplied from the battery 17.

  The EOPECU 20 includes a microcomputer 22 including a microprocessing unit. The microcomputer 22 generates a voltage signal Sv such as a PWM signal that controls driving of the inverter 21. The EOPECU 20 includes a pre-driver 23 that amplifies the voltage signal Sv. The pre-driver 23 generates voltage signals Sva to Svf obtained by amplifying the voltage signal Sv according to the switching elements 21 a to 21 f of the inverter 21.

  Further, the EOPECU 20 is provided with a current sensor 24 for detecting the current I flowing through the inverter 21. The current sensor 24 is provided between each switching element 21b, 21d, 21f on the downstream side of the inverter 21 and a reference potential point. The current I is input to a voltage signal generator 31 (to be described later) of the microcomputer 22.

Next, each function of the microcomputer 22 and the pre-driver 23 will be described in detail.
As shown in FIG. 2, the microcomputer 22 includes a switching control unit 30 that instructs to switch the energization pattern of the inverter 21 and a voltage signal generation unit 31 that generates the voltage signal Sv. The switching control unit 30 includes a magnetic pole detection unit (corresponding to a process in S36 described later) and an invalid period setting unit (described in S32 to be described later) that are realized by executing periodic processing based on a control program stored in a storage device (not illustrated). This corresponds to the processing of S35).

  The switching control unit 30 receives magnetic pole detection signals S <b> 1 to S <b> 3 from a terminal voltage detection unit 33 described later of the pre-driver 23. The magnetic pole detection signals S1 to S3 detect the magnetic pole position of the electric motor 10a, that is, the rotational position of the rotor 10b. And the switching control part 30 judges the timing which instruct | indicates switching the electricity supply pattern of the inverter 21 based on the detection result of a magnetic pole position. In this case, the switching control unit 30 generates a switching control signal Sc instructing to switch the energization pattern of the inverter 21 and outputs the switching control signal Sc to the voltage signal generating unit 31. The switching control unit 30 is connected to each phase 10u, 10v, 10w of the electric motor 10a. The switching control unit 30 receives the terminal voltages Vu, Vv, Vw of the phases 10u, 10v, 10w. In addition, each terminal voltage Vu, Vv, Vw as an analog signal is converted and input as a digital signal to the switching control unit 30 via a converter such as an A / D converter.

  The voltage signal generator 31 receives the electric pump control signal Seeop, the current I, and the switching control signal Sc. The voltage signal generation unit 31 generates a voltage signal Sv based on the electric pump control signal Seeop, the current I, and the switching control signal Sc, and outputs the voltage signal Sv to the voltage signal output unit 32 described later of the pre-driver 23.

  The voltage signal generator 31 calculates a target value indicating the current to be supplied to the electric motor 10a. And the voltage signal generation part 31 calculates | requires the deviation of the electric current I and target value, and performs electric current feedback (F / B) control so that this deviation may be eliminated. The voltage signal generator 31 generates the voltage signal Sv by executing current feedback (F / B) control.

As shown in FIG. 2, the pre-driver 23 includes a voltage signal output unit 32 that outputs voltage signals Sva to Svf and a terminal voltage detection unit 33 that generates magnetic pole detection signals S1 to S3.
The voltage signal output unit 32 receives the voltage signal Sv from the voltage signal generation unit 31 of the microcomputer 22. Based on the voltage signal Sv, the voltage signal output unit 32 generates voltage signals Sva to Svf obtained by amplifying the voltage signal Sv and applies them to the corresponding switching elements 21a to 21f.

  The terminal voltage detection part 33 is connected to each phase 10u, 10v, 10w of the electric motor 10a. The terminal voltage detector 33 receives the terminal voltages Vu, Vv, and Vw of the phases 10u, 10v, and 10w.

  Specifically, as shown in FIG. 3, the terminal voltage detection unit 33 includes three comparators 33u, 33v, and 33w. Corresponding terminal voltages Vu, Vv, and Vw are input to the non-inverting input terminals (“+”) of the comparators 33u, 33v, and 33w, respectively. Further, the midpoint voltage V0 (1/2 voltage of the power supply voltage Ve) of the power supply voltage Ve (voltage width) of the battery 17 is input to the inverting input terminals (“−”) of the comparators 33u, 33v, 33w, respectively. Is done. The midpoint voltage V0 is obtained by connecting the inverting input terminals of the comparators 33u, 33v, 33w between the resistors R1, R2 of the same value connected in series with the battery 17.

  Each of the comparators 33u, 33v, and 33w compares the analog signals of the terminal voltages Vu, Vv, and Vw with the midpoint voltage V0. Each comparator 33u, 33v, 33w outputs the result of this comparison as a binary digital signal of “1 (high level)” or “0 (low level)” as each magnetic pole detection signal S1 to S3. Each comparator 33u, 33v, 33w outputs "1" as each magnetic pole detection signal S1-S3, when each terminal voltage Vu, Vv, Vw is larger than the midpoint voltage V0. The comparators 33u, 33v, 33w output “0” as the magnetic pole detection signals S1 to S3 when the terminal voltages Vu, Vv, Vw are smaller than the midpoint voltage V0. When the magnitudes of the terminal voltages Vu, Vv, Vw and the midpoint voltage V0 are inverted, the magnetic pole detection signals S1 to S3 are changed from “1” → “0” (high / low) or “0” → “1”. (Low high) changes. The magnetic pole position of the electric motor 10a is detected by the change of each magnetic pole detection signal S1 to S3.

  Specifically, in the electric motor 10a, the energized phase is switched at every electrical angle of 60 ° through the switching of the energization pattern of each phase 10u, 10v, 10w by the inverter 21 (switching the combination of the energized phase and the non-energized phase) 120 °. Electric power is supplied by rectangular wave driving. That is, the energization of each phase 10u, 10v, and 10w is performed with an electrical angle of 360 °, a current-carrying section in the positive direction of the electrical angle of 120 ° (a direction in which the electric motor 10a rotates forward), and a non-energized section of the electrical angle of 60 °. It is executed by dividing into a current-carrying section in a negative direction with an electrical angle of 120 ° (direction in which the electric motor 10a rotates reversely) and a non-energized section with an electrical angle of 60 °. In the energization of the phases 10u, 10v, and 10w, the phases are shifted from each other by 120 ° in electrical angle.

  As shown in FIG. 4, the power supply voltage Ve appears in each terminal voltage Vu, Vv, Vw when energizing in the positive direction (hereinafter referred to as “positive energization”). In this case, “1” is output as each of the magnetic pole detection signals S1 to S3. Each terminal voltage Vu, Vv, Vw has a reference potential Vbase when it is energized in the negative direction (hereinafter referred to as “negative energization”). In this case, “0” is output as each of the magnetic pole detection signals S1 to S3.

  In this case, in each terminal voltage Vu, Vv, Vw, an induced voltage generated by the interlinkage magnetic flux of the stator 10c by the permanent magnet of the rotor 10b of the electric motor 10a appears when no power is supplied. Such an induced voltage exhibits a so-called zero-crossing change that intersects the midpoint voltage V0 with a predetermined gradient from the power supply voltage Ve or the reference potential Vbase. In this case, “1” → “0” (high / low) or “0” → “1” (low / high) is output as the magnetic pole detection signals S1 to S3.

  When the switching control unit 30 detects the change of “1” → “0” (high / low) or the change of “0” → “1” (low / high) by the magnetic pole detection signals S1 to S3, Detect the magnetic pole position. In this case, the switching control unit 30 determines that it is time to instruct switching of the energized phase.

  In addition, an edge-like induced voltage may appear during non-energization. Such an induced voltage expresses so-called spike noise that crosses the midpoint voltage V0 in an edge shape from the power supply voltage Ve or the reference potential Vbase. Spike noise appears due to the switching elements 21a to 21f switching from on to off. When spike noise appears, mask processing is performed to prevent the zero cross from being detected by the spike noise. The mask process will be described in detail later.

Next, a process executed when the electric motor 10a is driven will be described.
In the electric motor 10a, in the case of low rotation, the above-described zero cross (magnetic pole position) is difficult to detect in the non-energized phase. That is, the energized phase cannot be switched by detecting the magnetic pole position. Therefore, it is necessary to forcibly rotate the electric motor 10a during the start-up from the low rotation or stop of the electric motor 10a. The low rotation of the electric motor 10a means that the zero cross is rotating at a rotational speed that is empirically determined as being difficult to detect, and includes a so-called start-up from a stop. On the other hand, the high rotation of the electric motor 10a means that the electric motor 10a is rotating at a rotational speed that is empirically determined so that the zero cross can be detected, and includes driving at a rotational speed that is higher than a so-called steady rotational speed.

  The switching control unit 30 is a forced switching control for forcibly switching the energized phase when the electric motor 10a is started from a low rotation speed or from a stop, that is, a drive mode for determining (instructing) switching of the energized phase by so-called forced commutation. To control. In contrast, when the electric motor 10a rotates at a high speed, the switching control unit 30 determines (instructs) switching of the energized phase by so-called sensorless control, which is magnetic pole switching control that switches the energized phase based on detection of the magnetic pole position. To control.

  When the energized phase is forcibly switched by the forcible switching control, the switching control unit 30 switches the phase to be positively energized, negatively energized, or not energized every 8 ms. For example, when the phase 10u (U phase) is switched to the non-energized phase, the phase 10v (V phase) is switched to the energized phase (negatively energized), and the phase 10w (W phase) is switched to the energized phase (positive energized), The phase 10u is switched to the energized phase (positive energization), the phase 10v is switched to the energized phase (negative energization), and the phase 10w is switched to the non-energized phase. Next, the phase 10u is switched to the energized phase (positive energization), the phase 10v is switched to the non-energized phase, and the phase 10w is switched to the energized phase (negative energization). Next, the phase 10u is switched to the non-energized phase, the phase 10v is switched to the energized phase (positive energization), and the phase 10w is switched to the energized phase (negative energization). Next, the phase 10u is switched to the energized phase (negative energization), the phase 10v is switched to the energized phase (positive energization), and the phase 10w is switched to the non-energized phase. Next, the phase 10u is switched to the energized phase (negative energization), the phase 10v is switched to the non-energized phase, and the phase 10w is switched to the energized phase (positive energization). Switching in such a pattern is repeatedly executed while the forced switching control is executed.

  Then, the switching control unit 30 sets the drive mode to the magnetic pole when the rotational speed of the electric motor 10a increases to reach high rotation and the magnetic pole position of the electric motor 10a can be detected based on the induced voltage appearing in the non-energized phase. Shift to switching control. Thereafter, the switching control unit 30 controls the rotation of the electric motor 10a by magnetic pole switching control.

  Here, the magnetic pole switching control executed by the switching control unit 30 will be described. Note that the switching control unit 30 executes the following processing related to the magnetic pole switching control by executing periodic processing every control cycle (for example, 2 μs).

  Specifically, as shown in FIG. 5, the switching control unit 30 instructs to switch the energized phase based on the detection of the magnetic pole position (S31). In S31, switching control unit 30 instructs switching of the energized phase based on the same pattern as the above-described forced switching control.

  Next, the switching control unit 30 acquires a non-energized terminal voltage among the terminal voltages Vu, Vv, and Vw (S32). In S32, switching control unit 30 stores a digital value based on each terminal voltage Vu, Vv, Vw corresponding to the non-energized phase in a predetermined storage area as non-energized phase voltage value Vad. The non-energized phase voltage value Vad is a terminal voltage immediately after instructing to switch the energized phase based on detection of the magnetic pole position.

  Next, the switching control unit 30 determines whether or not the non-conduction phase voltage value Vad detected in S31 is larger than the power supply voltage Ve or smaller than the reference potential Vbase (S33). In S33, the switching control unit 30 determines whether or not the non-energized phase voltage value Vad is greater than the power supply voltage Ve when the non-energized phase is instructed to be switched from negative energization. On the other hand, the switching control unit 30 determines whether or not the non-energized phase voltage value Vad is smaller than the reference potential Vbase when the non-energized phase is instructed to switch from the positive energized phase. Note that the switching control unit 30 grasps whether the non-energized section is switched from negative energization or positive energization.

  As shown in FIG. 4, spike noise appears prior to detection of the magnetic pole position in the non-energized section. As enlarged in the figure, spike noise Nz (+) exceeding the power supply voltage Ve appears in the non-energization section switched from negative energization. While the spike noise Nz (+) appears, the corresponding terminal voltage also exceeds the power supply voltage Ve. That is, whether or not spike noise Nz (+) is appearing can be detected by determining whether or not the terminal voltage exceeds the power supply voltage Ve in the non-energized section switched from negative energization.

  On the other hand, as shown in an enlarged manner in FIG. 4, spike noise Nz (−) that is lower than the reference potential Vbase appears in the non-energization section that is switched from the positive energization. While the spike noise Nz (−) appears, the corresponding terminal voltage is also lower than the reference potential Vbase. That is, whether or not spike noise Nz (−) is appearing can be detected by determining whether or not the terminal voltage is lower than the reference potential Vbase in the non-energized section that is switched from positive energization.

  When the non-energized phase voltage value Vad detected in S32 is greater than the power supply voltage Ve in the non-energized section instructed to switch from negative energization (S33: YES), the switching control unit 30 spike noise Nz (+ ) Is appearing (S34). In addition, in the non-energization section instructed to switch from positive energization, the switching control unit 30 determines that the spike noise Nz (−) when the non-energization phase voltage value Vad detected in S32 is smaller than the reference potential Vbase (S33: YES). ) Is appearing (S34).

  Next, the switching control unit 30 performs a mask process on the spike noise determined in S32 (S35). In S35, the switching control unit 30 does not employ the magnetic pole detection signals S1 to S3 (deactivate even if input), for example, while determining that spike noise is appearing. Do not detect. After that, the switching control unit 30 returns to the process of S32 and repeatedly executes the processes of S32 to S35 until it can be determined that the spike noise is settled. That is, the switching control unit 30 sets a non-detection period (invalid period) in which the magnetic pole position is not detected by invalidating the detection results of the magnetic pole detection signals S1 to S3 for a predetermined period through the processes of S33 to S35.

  On the other hand, when the switching control unit 30 determines that the spike noise has been settled (S33: NO), whether or not the magnetic pole position is detected by the magnetic pole detection signal corresponding to the non-energized phase among the magnetic pole detection signals S1 to S3. Is determined (S36). At S36, switching control unit 30 determines whether or not there is a change of “1” → “0” (high / low) or “0” → “1” (low / high) based on the zero cross of the induced voltage appearing in the non-conduction phase. Judging. When the magnetic pole position is not detected (S36: NO), the switching control unit 30 repeatedly executes the process of S36 thereafter until the magnetic pole position is detected.

  On the other hand, when the magnetic pole position is detected (S36: YES), the switching control unit 30 returns to the process of S31 and instructs to switch the energized phase based on the magnetic pole position. In this case, the voltage signal generation unit 31 generates a voltage signal so as to switch the energized phase.

  Note that the switching control unit 30 forcibly switches after the electric motor 10a is temporarily stopped, for example, when the ignition of the vehicle is turned off, or when an abnormality (failure) is detected in the electric motor 10a or the inverter 21 that drives the electric motor 10a. Fail control such as restart from control is executed.

Next, the operation and effect of the EOPECU 20, which is a control device for the electric motor 10a of the present embodiment, will be described.
(1) Spike noise occurs when the electric motor 10a is rotating at a high speed, in the non-energized phase that is switched to non-energized, on the opposite side of the energized state up to that point (the direction of the reference potential Vbase with respect to positive energization, negative It is generated as a current that flows in the power supply voltage Ve) when energized. As described above, the current flowing on the opposite side in this case flows larger than usual, so that the width of the voltage fluctuation appears larger than the power supply voltage as the power supply source of the motor. That is, the spike noise is characterized in that a voltage larger than the power supply voltage or a voltage smaller than the reference potential is instantaneously generated in the non-energized phase switched from energized to non-energized. This embodiment utilizes such a feature of spike noise.

  That is, according to the present embodiment, when the electric motor 10a rotates at high speed, the processes of S32 and S33 are executed, so that the induced voltage that appears in the non-conduction phase, that is, the phenomenon of spike noise that occurs in the non-conduction phase itself. Can be detected. Therefore, the non-detection period can be set based on whether or not spike noise occurs.

  For example, as shown in FIG. 6A, spike noise generated in the non-energized phase is detected in S32 and S33. While the spike noise is detected in this case, it is a non-detection period in which the zero cross is not detected by the magnetic pole detection signals S1 to S3 in S34 and S35.

  Further, as shown in FIG. 6B, spike noise having a longer period than that in FIG. 6A is detected in S32 and S33 as described above. While spike noise is detected in this case, it is a non-detection period during which no zero cross is detected by the magnetic pole detection signals S1 to S3 in S34 and S35, as described above.

  Further, as shown in FIG. 6C, spike noise of a short period compared to FIGS. 6A and 6B is detected in S32 and S33 as described above. While spike noise is detected in this case, it is a non-detection period during which no zero cross is detected by the magnetic pole detection signals S1 to S3 in S34 and S35, as described above.

  That is, when the electric motor 10a rotates at a high speed, the non-detection period varies depending on the period in which the spike noise occurs even if the period in which the spike noise occurs changes. As a result, it is possible to cope with spike noise that appears without restriction due to the length of the generated period. Therefore, erroneous detection of the magnetic pole position related to spike noise can be suppressed, and the occurrence of lock and step-out of the electric motor 10a can be suppressed.

  (2) Specifically, the non-detection period is set to a period in which the non-energized phase voltage value Vad is larger than the power supply voltage Ve when it is instructed to switch from the negative energized phase to the non-energized phase. In contrast, the non-detection period is set to a period in which the non-energized phase voltage value Vad is smaller than the reference potential Vbase when it is instructed to switch the non-energized phase from positive energization.

  That is, this embodiment can exhibit an effect only when the feature of spike noise appears when the electric motor 10a rotates at high speed. Therefore, the period for setting the non-detection period can be optimized.

In addition, you may change the said embodiment as follows.
In S33 of FIG. 5, it may be determined whether or not the power supply voltage Ve is equal to or higher than the reference potential Vbase. The threshold value set here may be set to be less than the power supply voltage Ve or greater than the reference potential Vbase in consideration of the fluctuation of the power supply voltage Ve. Further, the threshold value may be set with a width.

  In S33 of FIG. 5, it is determined that spike noise has occurred due to a predetermined change in the induced voltage (simple drop, gradient, etc.) due to switching to the non-energized phase, and a subsequent predetermined change in the induced voltage (simply It may be determined that the spike noise is settled by a head or a gradient.

  In the magnetic pole switching control, spike noise may be generated in principle after the energized phase is switched, and it may be determined in S33 in FIG. 5 that only spike noise is contained.

  In S33 of FIG. 5, different processing may be executed depending on whether the non-energized phase is switched from positive energization or negative energization. In this case, prior to S33, it may be determined whether the non-energized phase is switched from positive energization or negative energization.

  In S35 of FIG. 5, the switching control unit 30 does not input the magnetic pole detection signals S1 to S3 themselves during this period, that is, stops the magnetic pole position detection function (does not detect the magnetic pole position). Good. In this case, when the non-energized phase voltage value Vad is larger than the power supply voltage Ve or the non-energized phase voltage value Vad is smaller than the reference potential Vbase, the magnetic pole detection signals S1 to S3 are output in the terminal voltage detection unit 33. It is only necessary to incorporate a circuit that cannot be used.

  The terminal voltage detection unit 33 may be incorporated in the microcomputer 22. Further, the EOPECU 20 does not need to include the pre-driver 23 and may cause the microcomputer 22 to have the function of the pre-driver 23.

  In the above embodiment, while the forced switching control is executed, the energized phase is forcibly switched every 8 ms. However, the method for forcibly switching the energized phase may be appropriately changed. For example, a method of detecting the magnetic pole position in advance while the electric motor 10a is stopped and forcibly switching the energized phase based on the result may be used.

  -In above-mentioned embodiment, although the supply destination of the oil of the electric pump 10 was CVT11, a stepped transmission may be sufficient. Further, the supply destination of the oil of the electric pump 10 may be the internal combustion engine 12. Moreover, the electric pump 10 may supply the refrigerant | coolant for cooling the secondary battery (battery) mounted in a hybrid vehicle, an electric vehicle, etc., for example.

Next, a technical idea that can be grasped from the above embodiment and its modifications will be additionally described.
(A) In the electric motor control device, the magnetic pole position detection unit does not detect the magnetic pole position by the invalid period setting unit when the energization state of the three-phase motor coil is switched by the switching control unit. The magnetic pole position is detected on condition that the period is canceled. According to the above configuration, the processing flow of magnetic pole detection can be optimized and embodied.

  (B) In the control device for the electric motor, the switching control unit performs the magnetic pole switching control when the electric motor is in a steady rotation after starting from the stop, and the invalid period setting unit is controlled by the switching control unit. The invalid period can be set when the magnetic pole switching control is performed. According to the above configuration, the invalid period setting can be optimized and embodied.

  Nz (+), Nz (-) ... spike noise, Vad ... non-energized phase voltage value, Vu, Vv, Vw ... terminal voltage, S1 to S3 ... magnetic pole detection signal, 10a ... electric motor, 10b ... rotor, 10c ... stator 10u, 10v, 10w ... three-phase motor coils, 20 ... EOPECU, 22 ... microcomputer, 30 ... switching control unit (magnetic pole position detection unit, invalid period setting unit).

Claims (3)

Of the three-phase motor coils of the electric motor having a stator having a three-phase motor coil and a rotor that forms a field, two phases are energized and the remaining one is a non-energized phase. In the control device for the electric motor that controls the control amount of the electric motor by switching the energization state of the motor coil,
A magnetic pole position detector that detects the magnetic pole position of the rotor based on an induced voltage that appears in the non-energized phase by energizing the energized phase;
A switching control unit that executes magnetic pole switching control for switching the energization state of the three-phase motor coil based on detection of the magnetic pole position of the rotor detected by the magnetic pole position detection unit;
When the magnetic pole switching control is executed by the switching controller, an invalid period is set to invalidate the detection result of the magnetic pole position detector for a predetermined period prior to the detection of the magnetic pole position by the magnetic pole position detector. A period setting section;
An electric motor control device comprising:   The invalid period setting unit sets the invalid period while an induced voltage appearing in the non-energized phase is larger than a power supply voltage or while an induced voltage appearing in the non-energized phase is smaller than a reference potential. The control apparatus of the electric motor of description. The switching control unit performs switching so that positive direction energization, negative direction energization, and non-energization appear one phase at a time for the energization state of the three-phase motor coil,
The invalid period setting unit
For the non-energized phase after switching from energization in the negative direction to the non-energized phase, a period in which the induced voltage appearing in the non-energized phase is larger than the power supply voltage is set as the invalid period,
The electric motor according to claim 2, wherein a period in which an induced voltage appearing in the non-energized phase is smaller than a reference potential is set as the invalid period for the non-energized phase after switching from the positive energization to the non-energized. Control device.