JP5402403B2 - Electric motor control system - Google Patents

Description

  The present invention relates to an electric motor control system that controls an electric motor driven by multiphase AC power.

  2. Description of the Related Art Conventionally, an electric motor control system that controls output torque of an electric motor by controlling a switching operation of power conversion means such as an inverter is known. If one of the switching elements constituting the power conversion means has an open failure or one of the wires connecting the motor and the inverter is disconnected (open phase failure), no current can flow through the failure phase. In this case, it is necessary to drive the motor in the remaining phase and it is possible to continue running.However, since the motor torque includes a pulsation component, the rotation is not smooth and weak vibration is generated. May occur.

  For example, Patent Literature 1 discloses a technique for detecting the occurrence of a failure in a power conversion device. According to such a method, the current change rates of the d-axis current and the q-axis current on the rotation coordinates are calculated, and when the change rate exceeds a set value, an open phase detection signal is output.

JP 2003-348898 A

  However, according to the method disclosed in Patent Document 1, although it is possible to detect that a failure has occurred in some phase, it is necessary to identify a specific failure location where the failed phase is. There is a problem that can not be.

  The present invention has been made in view of such circumstances, and an object of the present invention is to detect a failure of the power conversion means including a failure point.

  In order to solve such a problem, the present invention detects the d-axis current and the q-axis current of the electric motor based on the phase current of each phase detected by the current detection means, and detects the detected d-axis current or It is detected that the q-axis current is in a zero state. Then, when a zero state is detected, failure detection that identifies a failure location in the power conversion unit is performed based on the rotation angle of the electric motor detected by the rotation angle detection unit.

  According to the present invention, since there is a certain pattern according to the failure location in the combination of the zero state of the d-axis current or the q-axis current and the rotation angle of the motor at that time, these factors are taken into consideration. As a result, the failure location of the power conversion means can be specified.

Explanatory drawing which shows typically the whole structure of the control system concerning 1st Embodiment Illustration of instantaneous space current vector at normal time when the motor is driven by a three-phase inverter Explanatory drawing which shows the locus | trajectory of the current vector at the time of U arm upper arm UP failure (for example, open failure) Explanatory drawing which shows the locus | trajectory of electric current vector I 'when each arm has an open failure during power running and forward rotation The block block diagram at the time of grasping functionally the control apparatus 30 concerning 1st Embodiment Explanatory drawing showing a list of electrical angles when the current vector locus is orthogonal to the d-axis or q-axis Explanatory drawing which shows transition of dq axis current Id and Iq, and electrical angle (theta) of the motor 10. FIG. The block block diagram at the time of grasping functionally the control apparatus 30 concerning 2nd Embodiment

(First embodiment)
FIG. 1 is an explanatory diagram schematically showing the overall configuration of the control system according to the first embodiment of the present invention. In the present embodiment, a motor control system applied to a drive motor for an electric vehicle (for example, an electric vehicle) will be described. This motor control system is mainly composed of a motor 10, an inverter 20, and a control device 30.

  The motor 10 is mainly composed of a rotor and a stator, and has a plurality of phase windings (in this embodiment, a U-phase winding, a V-phase winding, and a W-phase) that are star-connected around a neutral point. This is a permanent magnet synchronous motor in which three phase windings comprising windings are wound around a stator. The motor 10 is driven by the interaction between a magnetic field generated when a three-phase alternating current is supplied to each phase winding from the inverter 20 and a magnetic field generated by a permanent magnet of the rotor. Thereby, a rotor and the output shaft connected with this rotate. The output shaft of the motor 10 is connected to, for example, an automatic transmission of an electric vehicle.

  The inverter 20 is connected to the battery 21, and is power conversion means that converts DC power from the battery 21 into AC power and supplies the AC power to the motor 10. The AC power is generated corresponding to each phase of the motor 10, and the AC power of each phase is supplied to the motor 10. Here, the battery 21 is configured to be able to store a predetermined capacity of electric power, and functions as a DC power source. As the battery 21, for example, a nickel metal hydride battery or a lithium ion battery can be used.

  The inverter 20 is a circuit in which an upper arm (switching means) connected to the bus on the positive side of the battery 21 and a lower arm (switching means) connected to the bus on the negative side of the battery 21 are connected in series. Corresponding to each phase of the phase, V phase and W phase. In FIG. 1, UP, VP, and WP indicate the upper arms corresponding to the U phase, V phase, and W phase, and UN, VN, and WN correspond to the U phase, V phase, and W phase, respectively. The lower arm is shown. The upper arms UP to WP and the lower arms UN to WN are mainly composed of a semiconductor switch 22 (for example, a transistor such as an IGBT) that can control conduction in one direction. The diode 23 is connected in reverse parallel. The on / off state of each arm UP to WN, that is, the on / off state (switching operation) of the semiconductor switch 22 is controlled through a PWM signal output from the control device 30. The semiconductor switches 22 constituting the individual arms UP to WN are turned on when turned on through the PWM signal of the control device 30 and turned off (cut off) when turned off.

  The control device 30 controls the output torque of the motor 10 by controlling the switching operation of each arm UP to WN of the inverter 20. As the control device 30, a microcomputer mainly composed of a CPU, a ROM, a RAM, and an I / O interface can be used. The control device 30 performs a calculation for controlling the inverter 20 in accordance with a control program stored in the ROM. Then, control device 30 outputs a control signal (PWM signal) calculated by this calculation to inverter 20.

  The control device 30 controls the switching operation of the inverter 20, specifically, the on / off states of the individual semiconductor switches 22 corresponding to the upper and lower arms UP to WN, for each phase, by a control method such as PWM wave voltage driving. The PWM wave voltage drive generates a PWM wave voltage from a DC voltage and applies it to the motor 10. Specifically, PWM control is performed based on the carrier voltage and the sine wave control voltage, and the duty command value for PWM control is set. This is a driving method in which an equivalent sine wave AC voltage is applied to the motor 10 by calculation. The control device 30 calculates a sinusoidal control voltage for each phase based on the torque command given from the outside, the rotation speed (electrical angular velocity) of the motor 10 and the current value of each phase, and thereby the switching of the inverter 20 is performed. Control the behavior.

  Here, sensor signals from various sensors are input to the control device 30. The current sensor 11 detects U-phase and V-phase currents Iu and Iv in the motor 10, respectively. Since the three-phase current has a relationship that the sum is zero, the control device 30 can detect the remaining W-phase current based on the U-phase and V-phase currents Iu and Iv. The rotation angle sensor 12 is, for example, a resolver or a rotary encoder, and detects the electrical phase θ (electric angle) of the motor 10, that is, the rotation angle of the motor 10, based on position information representing the rotor position of the motor 10. doing.

  Further, as one of the features of the present embodiment, the control device 30 is fixed together with the above-described normal motor control while the switching element is open (off) in any arm UP to WN of the inverter 20. A so-called open failure or a so-called open-phase failure in which one of the electric wires of each phase connecting between the motor 10 and the inverter 20 is disconnected is detected (failure detection). The concept of failure detection according to the present embodiment will be described below prior to description of specific processing contents of failure detection by the control device 30.

FIG. 2 is an explanatory diagram of an instantaneous space current vector in a normal state when the motor is driven by a three-phase inverter. Here, currents (instantaneous values) flowing through the U-phase, V-phase, and W-phase are Iu, Iv, and Iw. As shown in FIG. 2, a two-phase current coordinate system (αβ coordinate system) obtained by coordinate transformation of the three-phase current is placed under the condition that the three-phase current sum is zero. When the α axis is the real axis and the β axis is the imaginary axis, the instantaneous space current vector I ′ is expressed by the following equation on the complex plane.

  When the motor is rotating at a constant torque, the current vector I 'is a vector of magnitude | I | at an angle θ from the U-phase current basic vector. θ is equal to the electrical angle of the motor, and the motor rotates at the electrical angular velocity ω (θ = ωt (t: time)). In this case, the current vector I 'rotates at the same angular velocity on the UVW coordinate axis, and its locus draws a circle. The dq coordinate axis, which is a coordinate on the rotor, rotates in synchronization with θ, and the current vector I ′ is stationary with respect to the dq coordinate axis and becomes a fixed vector.

  Here, the dq-axis coordinate system is an orthogonal coordinate system including a d-axis and a q-axis that rotate at an electrical rotation speed that is an integral multiple of the mechanical rotation speed of the motor. In the three-phase synchronous motor, the dq axis coordinate system rotates in synchronization with the motor rotation. With the dq axis coordinate system, the current supplied to the stator winding of the motor is divided into a field component current (d axis current) and a torque component current (q axis current) and displayed as a vector.

  FIG. 3 is an explanatory diagram showing the locus of the current vector when the upper arm UP of the U phase has failed (for example, open failure). Here, it is assumed that the upper arm UP of the U phase has an open failure. Factors that lead to an open failure include destruction due to overvoltage / overcurrent of a switching element body such as an IGBT forming the upper arm UP, failure of a drive circuit of the switching element, and the like. Here, the case where the sign of the U-phase current Iu flows from the inverter to the motor is positive. Further, it is assumed that the U-phase current Iu is negative immediately before the upper arm UP has an open failure (Iu <0 (a scene where current flows from the motor to the inverter).

  Even after an open failure of the switching element of the upper arm UP, by switching the freewheeling diode of the upper arm UP or the switching element and freewheeling diode of the U-phase lower arm UN, the U-phase current Iu is the same as normal. It becomes possible to flow. For this reason, immediately after the open failure, as in FIG. 2, the instantaneous current vector I 'rotates with respect to the UVW phase stationary coordinates while synchronizing with the motor for a while, and the tip locus draws an arc. In this case, the U-phase current Iu becomes a sine wave with respect to time.

It is assumed that the rotation of the motor has progressed to a point where the U-phase current Iu rises in the positive direction across 0A. However, at this time, since the switching element of the upper arm UN has an open failure because the switching element of the upper arm UN cannot be turned on, the DC voltage of the power source cannot be applied to the U-phase coil of the motor. As a result, the U-phase current Iu sticks in the vicinity of 0 A, and cannot increase further. In this case, the sum of the V-phase current Iv and the W-phase current Iw becomes zero under the condition that the sum of the three-phase currents of the motor is balanced. Therefore, Iv = −Iw, and the spatial current vector I ′ is expressed by the following equation according to Equation 1.

  As can be seen from the equation, the current vector I ′ has only an imaginary component and only a β-axis component. Therefore, the locus of the current vector I ′ moves linearly on the β axis as the rotor rotates. Therefore, as the rotation proceeds, the current vector I 'moves from the point a in FIG. 3 to the origin (point b) and the point c on the β axis. In the vicinity of the point c, the switching element of the U-phase lower arm UN is turned on / off again to apply a voltage to the U-phase coil, to flow a negative U-phase current, and to set the current vector I ′ to α It becomes possible to make the region <0. As a result, the current vector I ′ is separated from the β axis again and rotates around the origin. As a result, the motor moves to point c, point d, and point a in FIG. Then, after moving to the point a, the U-phase current Iu sticks to 0 A again, the locus of the current vector I ′ starts to move on the β axis, and the above-described state is repeated.

  Thus, the locus of the current vector I 'when the U phase has an open failure is composed of a semicircular curved locus in which the motor current flows in three phases and a linear locus in which two phases of the UW phase flow.

  Here, a power running positive rotation state in which the q-axis current Iq, which is the q-axis component of the current vector I ′, is positive and the electrical angular velocity ω is positive is considered. In this case, if the motor is controlled in a state where the d-axis current Id is zero, the current vector I ′ is on the β-axis in a scene where the electrical angle θ is 180 ° to 360 °. When the d-axis current Id, which is the d-axis component of the current vector I ′, is performing negative field weakening control, the current vector I ′ is on the β axis because the electrical angle θ is 180 ° −δ to 360 It is a scene of ° + δ. Here, δ is the current vector phase angle on the dq coordinate axis when the operation is normal (Id = Iasin δ, Iq = iacos δ).

  Consider a case where the current vector I ′ is on the β axis when the electrical angle θ is 270 °. At this time, the d-axis vector faces the negative direction of the β axis, and the q-axis vector faces the positive direction of the α axis and is orthogonal to the β axis. Since the current vector I ′ is on the β axis and orthogonal to the q axis, the q axis current Iq at this time becomes zero. Similarly, when the electrical angle θ is 180 °, the d-axis current Id becomes zero because the d-axis and the current vector I ′ are orthogonal to each other.

  Thus, it can be seen that when the upper arm UP of the U phase fails, the q-axis current becomes 0 when the electrical angle θ is 270 °, and the d-axis current becomes 0 when the electrical angle θ is 180 °. As shown in FIG. 2, this is controlled when the current vector I ′ is controlled to draw an arc during normal operation, that is, when the dq axis current is controlled to be constant. It does not depend on the method.

  FIG. 4 is an explanatory diagram showing the locus of the current vector I ′ when each arm has an open failure during power running / forward rotation. When any of the arms UP to WN has an open failure, the locus of the current vector I 'is composed of a semicircular curved locus and a straight locus, as in FIG. At this time, if the fault location is different, the electrical angle range in which the current vector I 'becomes a straight line changes. For example, when the U-phase lower arm UN has an open failure, the locus of the current vector I ′ is linear when the electrical angle θ is 0 ° to 180 °. The q-axis current is zero because the q-axis and the current vector locus are orthogonal when the electrical angle θ is 90 °.

  Similarly, when the V-phase upper arm VP has an open failure, the locus of the current vector I ′ becomes linear when the electrical angle θ is 300 ° to 120 °. The q-axis current is zero because the q-axis and the current vector locus are orthogonal when the electrical angle θ is 30 °. In addition, when the V-phase lower arm VN has an open failure, the locus of the current vector I ′ becomes linear when the electrical angle θ is 120 ° to 300 °. The q-axis current is zero because the q-axis and the current vector locus are orthogonal when the electrical angle θ is 210 °.

  Further, when the W-phase upper arm WP has an open failure, the locus of the current vector I ′ becomes linear when the electrical angle θ is 60 ° to 240 °. The q-axis current is zero because the q-axis and the current vector locus are orthogonal when the electrical angle θ is 30 °. In addition, when the V-phase lower arm VN has an open failure, the locus of the current vector I ′ becomes linear when the electrical angle θ is 120 ° to 300 °. The q-axis current is zero because the q-axis and the current vector locus are orthogonal when the electrical angle θ is 210 °.

  In this manner, since the current vector I ′ has a unique trajectory corresponding to each failure location, the angle at which the dq axis is orthogonal to the trajectory of the current vector I ′ also varies depending on the failure location. Therefore, if it is known that the dq-axis current is close to zero at a specific electrical angle, it becomes possible to specify the phase of the inverter open failure.

  FIG. 5 is a block configuration diagram when the control device 30 according to the first embodiment is functionally grasped. Hereinafter, a failure detection process performed by the control device 30 according to the present embodiment will be described. In relation to failure detection which is one of the features of the present embodiment, the control device 30, when functionally grasping this, the three-phase to two-phase converter 31, the d-axis current zero-cross detector 32, q The shaft current zero-cross detection unit 33 and the abnormality determination unit 34 are provided.

The three-phase to two-phase converter 31 is obtained from the three-phase currents Iu to Iw (substantially, the two-phase currents Iu and Iv of the three phases) detected by the current sensor 11 and the rotation angle sensor 12. Based on the electrical angle θ, coordinate conversion is performed using the following equation to calculate the dq-axis currents Id and Iq.

  The calculated d-axis current Id is output to the d-axis current zero-cross detector 32, and the q-axis current Iq is output to the q-axis current zero-cross detector 33.

  Each of the d-axis current zero-cross detection unit 32 and the q-axis current zero-cross detection unit 33 detects that the input d-axis current Id or the q-axis current Iq is in a zero state, and abnormally detects the detection flags Fd and Fq at the time of detection. It outputs to the determination part 34.

  As a configuration of the dq-axis current zero-cross detection units 32 and 33, for example, a method of detecting that the absolute values of the input dq-axis currents Id and Iq are smaller than a predetermined determination threshold is used. That is, the dq-axis current zero cross detection units 32 and 33 not only detect that the dq-axis currents Id and Iq are strictly zero, but also detect a range that can be regarded as almost zero as a zero state. In this case, the determination threshold value is a value in a range in which the dq-axis currents Id and Iq can be regarded as substantially zero on the basis of the above-described control concept in consideration of variations in the current sensor 11 and a current sampling period. It can be selected as a determination threshold. Further, the determination threshold value may be set so as to have an appropriate margin so that the zero state is not erroneously detected due to noise included in the detection values of the current sensor 11 and the rotation angle sensor 12.

  When the detection flags Fd and Fq are output from the dq-axis current zero-cross detection units 32 and 33, the abnormality determination unit 34 performs failure detection based on the electrical angle θ of the motor 10 read from the rotation angle sensor 12. Specifically, as described with reference to FIGS. 3 and 4, the abnormality determination unit 34 determines that the electrical angle θ corresponding to the output of the detection flags Fd and Fq is when the current vector locus is orthogonal to the d axis or the q axis. It is determined whether it is an electrical angle. Thereby, the abnormality determination unit 34 identifies the occurrence of the failure and the location of the failure. Here, as shown in FIG. 6, the abnormality determination unit 34 holds a list (map) of electrical angles in order to identify a failure site, and performs failure detection based on the list. Here, FIG. 6 is an explanatory diagram showing a list of electrical angles when the current vector locus is orthogonal to the d-axis or q-axis.

  It should be noted that the current electrical angle θ is within a specific angle range as well as the electrical angle when the current vector locus at the time of abnormality is orthogonal to the d-axis or q-axis and the current electrical angle θ. A failure may be detected under the conditions. In this case, the angle range used for determination includes the electrical angle when the current vector locus at the time of abnormality is orthogonal to the d-axis or q-axis, and further takes into account variations, errors, sample period, and sensor noise of the rotation angle sensor 12. It is preferable to set so as to include an appropriate margin.

  FIG. 7 is an explanatory diagram showing transitions of the dq-axis currents Id and Iq and the electric angle θ of the motor 10. In the same figure, (a) shows the transition in the normal state, (b) shows the transition of the U-phase upper arm UP, (c) shows the transition of the V-phase upper arm VP, and (d) shows the transition under the U-phase. The transition of the arm UN is shown. The data shown in the figure shows a state in which the motor 10 is operated by power running and forward rotation while performing field control while the d-axis current Id is negative and the q-axis current Iq is positive and weakened. Further, in the figure, it is set so that a failure occurs after a lapse of a certain time (time A) from the start of the motor operation.

  As shown in the figure, the dq-axis currents Id and Iq operate at a constant value before the occurrence of the failure, but all the waveforms fluctuate greatly after the occurrence of the failure. For example, as shown in (b), when the U-phase upper arm UP fails, the electrical angle θ is 180 ° and the d-axis current Id is zero, and the electrical angle θ is 270 ° and the q-axis current Iq is zero. It becomes. When the V-phase upper arm VP fails, the electrical angle θ is 300 ° and the d-axis current Id is in the zero state, and the electrical angle θ is 30 ° and the q-axis current Iq is in the zero state. Further, when the U-phase lower arm UN fails, the d-axis current Id becomes zero when the electrical angle θ is 0 °, and the q-axis current Iq becomes zero when the electrical angle θ is 90 °.

  These states coincide with the angles at which the current vector locus and the dq axis shown in FIG. 6 are orthogonal to each other. In this manner, the abnormality determination unit 34 can detect the failure location based on the electrical angle θ when the dq-axis current zero-cross detection units 32 and 33 detect the zero state.

  As described above, in the present embodiment, according to the control device 30, the three-phase to two-phase conversion unit 31 is based on the phase currents Iu to Iw of each phase detected by the current sensor 11 (phase current detection means). Ten d-axis currents Id and q-axis current Iq are detected (dq-axis current detection means). Further, the dq axis current zero cross detectors 32 and 33 detect that the d axis current Id or the q axis current Iq is in a zero state (zero state detecting means). Further, the abnormality determination unit 34 is based on the rotation angle θ of the motor 10 detected by the rotation angle sensor 12 (rotation angle detection means) when the zero state of the d-axis current Id or the q-axis current Iq is detected. Then, the failure detection for specifying the failure location in the inverter 20 is performed (identification means).

  According to such a configuration, not only a failure has occurred in the inverter 20 based on the zero state of the d-axis current Id or the q-axis current Iq and the rotation angle θ of the motor 10 at that time, but also the failure location is included. Can be detected. Further, according to the conventional method, in order to detect a failure once, it is necessary for the motor to reliably perform one or more rotations. On the other hand, according to the present embodiment, a failure can be detected simply by determining whether or not the current when the electrical angle (rotation angle) θ of the motor 10 passes a predetermined angle is within a specific range. Failure detection can be performed even if the motor 10 is less than one rotation. If field-weakening control is performed and the d-axis current Id is not zero, two failures are detected at the timing when the d-axis current Id becomes zero and the q-axis current Iq becomes zero during one rotation of the motor 10. There is an opportunity to do. Therefore, failure detection can be performed quickly and accurately even in a scene where the motor 10 rotates at a low speed.

  Further, according to the present embodiment, the abnormality determination unit 34 provides a map that defines the correspondence relationship between the electrical angle θ of the motor 10 and the failure location when the d-axis current Id or the q-axis current Iq becomes zero. The failure location in the inverter 20 is specified with reference. According to such a configuration, the failure location can be specified by referring to the map, so that the failure detection can be realized with a simple configuration.

  In the present embodiment, the dq-axis current zero-cross detection units 32 and 33 detect the absolute value of the detected d-axis current Id or q-axis current Iq smaller than a predetermined determination threshold value for determining the zero state. Detect zero condition. Further, the abnormality determination unit 34 corresponds to a failure corresponding to this map when the rotation angle θ of the motor 10 detected by the rotation angle sensor 12 is included in a predetermined angle range including the rotation angle of the electric motor described in the map. Identify the location. According to this configuration, failure detection can be performed in consideration of variations in the current sensor 11 and current sampling cycle, variations in the rotation angle sensor 12, errors, sampling cycle, and sensor noise. The detection accuracy can be improved.

(Second Embodiment)
FIG. 8 is a block configuration diagram when the control device 30 according to the second embodiment is functionally grasped. The control device 30 according to the present embodiment is different from that of the first embodiment in that internal processing for suppressing erroneous detection of a failure is added. Note that description of points that are the same as in the first embodiment will be omitted, and hereinafter, differences will be mainly described.

  As shown in FIG. 8, when this is functionally grasped, the control device 30 has a three-phase to two-phase converter 31, a d-axis current zero-cross detector 32, a q-axis current zero-cross detector 33, and a current target. A value generation unit 35, a current prediction unit 36, and an abnormality determination unit 37 are included. This control device 30 is configured not to erroneously detect a failure when the dq-axis currents Id and Iq become zero even though the inverter 20 is operating normally.

  The three-phase to two-phase converter 31 is obtained from the three-phase currents Iu to Iw (substantially, the two-phase currents Iu and Iv of the three phases) detected by the current sensor 11 and the rotation angle sensor 12. Based on the electrical angle θ, coordinate conversion is performed using the calculation formula shown in Formula 3, and the dq-axis currents Id and Iq are calculated. The calculated d-axis current Id is output to the d-axis current zero-cross detector 32, and the q-axis current Iq is output to the q-axis current zero-cross detector 33.

  Each of the d-axis current zero-cross detection unit 32 and the q-axis current zero-cross detection unit 33 detects the zero state of the input d-axis current Id or the q-axis current Iq, and detects the detection flags Fd and Fq at the time of detection. Output to.

  The current target value generation unit 35 calculates dq-axis current target values Id * and Iq * for controlling the switching operation of the inverter 20 so that the motor 10 generates necessary torque. For example, the current target value generation unit 35 holds a map that describes the correspondence between the torque command value required for the motor 10 and the electrical angular velocity ω of the motor 10 and the dq-axis current target values Id * and Iq *. Based on the torque command value and the electrical angular velocity ω, dq-axis current target values Id * and Iq * are calculated. The calculated dq-axis current target values Id * and Iq * are output to the current prediction unit 36.

  Based on the dq-axis current target values Id * and Iq *, the current prediction unit 36 sets the dq-axis current when the inverter 20 operates normally without failure as non-failure dq-axis current predicted values Idc and Iqc. Predict. For example, when the motor 10 is controlled using PI control and voltage non-interference control, the response waveform of the dq axis current is considered to be a first-order lag system. Therefore, the current prediction unit 36 of the present embodiment inputs the dq-axis current target values Id * and Iq * to the low-pass filter represented by “ωc / (s + ωc)”, so that the predicted dq-axis current values Idc and Iqc. Is calculated. The predicted dq-axis current predicted values Idc and Iqc are output to the abnormality determination unit 37.

  When the detection flags Fd and Fq are output from the dq-axis current zero-cross detection units 32 and 33, the abnormality determination unit 37 performs failure detection based on the electrical angle θ of the motor 10 read from the rotation angle sensor 12. Specifically, as described with reference to FIGS. 3 and 4, the abnormality determination unit 34 determines that the electrical angle θ corresponding to the output of the detection flags Fd and Fq is when the current vector locus is orthogonal to the d axis or the q axis. It is determined whether it is an electrical angle. Thereby, the abnormality determination unit 34 identifies the occurrence of the failure and the location of the failure. Here, as shown in FIG. 6, the abnormality determination unit 34 holds a list of electrical angles in order to identify a failure site, and performs failure detection based on the list.

  In the present embodiment, the abnormality determination unit 37 does not perform failure determination when the absolute values of the dq-axis current predicted values Idc and Iqc are smaller than the threshold ε. That is, the abnormality determination unit 37 performs failure detection by the above-described method when the dq-axis current predicted values Idc and Iqc are equal to or greater than the threshold ε.

  As described above, in the present embodiment, the control device 30 is based on the d-axis current command value Id * and the q-axis current command value Iq * specified from the torque command required for the motor 10 when the inverter 20 is not in failure. Are predicted as d-axis current predicted values Idc and Iqc (prediction means). In this case, the abnormality determination unit 34 does not detect a failure when the d-axis current predicted value Idc or the q-axis current predicted value Iqc is in a zero state.

  According to such a configuration, when the dq-axis current prediction values Idc and Iqc are near zero, the failure detection is masked, so that when the inverter 20 is in a normal state, the dq-axis currents Id and Iq are erroneously detected as being near zero. Detection can be suppressed. For example, when the torque is changed from positive to negative in order to change the motor 10 from the power running state to the regenerative state, the dq axis currents Id and Iq are transiently zero even if the inverter 20 has not failed. May take a nearby value. If the electrical angle θ when the dq-axis currents Id and Iq are close to zero overlaps with the angle shown in FIG. 6, there is a possibility that a failure is erroneously detected. However, according to the present embodiment, failure detection can be performed effectively without causing such inconvenience.

  In the present embodiment, the dq-axis current predicted values Idc and Iqc are obtained from the dq-axis current target values Id * and Iq * by the primary low-pass filter. However, the prediction method is not limited to this, and various methods are available. It is possible to use a technique.

  In each of the above-described embodiments, the case where the motor 10 is rotated forward has been described. However, even when the motor 10 rotates in the reverse direction, it is possible to detect the occurrence of a failure and the location of the failure based on the same concept. In each of the above-described embodiments, the failure detection in the permanent magnet synchronous motor and the three-phase inverter has been described. However, the present invention is not limited to this, and a multi-phase inverter or other AC motor (winding type) is used. A combination of a synchronous motor, an induction motor, etc.) can detect a failure with the same configuration.

DESCRIPTION OF SYMBOLS 10 ... Motor 11 ... Current sensor 12 ... Rotation angle sensor 20 ... Inverter 21 ... Battery 22 ... Semiconductor switch 23 ... Recirculation diode 30 ... Control device 31 ... Three-phase two-phase conversion part 32 ... d-axis current zero cross detection part 33 ... q Axis current zero cross detection unit 34 ... abnormality determination unit 35 ... current target value generation unit 36 ... current prediction unit 37 ... abnormality determination unit

Claims (4)

In the motor control system that controls the motor driven by the polyphase AC power,
A circuit in which an upper arm connected to the positive electrode side of the power supply and a lower arm connected to the negative electrode side of the power supply are connected in series is provided corresponding to each phase of the electric motor, and DC power from the power supply is phased. Power conversion means for converting into alternating current power and outputting to the electric motor;
Rotation angle detection means for detecting the rotation angle of the electric motor;
Phase current detection means for detecting a phase current of each phase of the electric motor;
Dq-axis current detection means for respectively detecting the d-axis current and the q-axis current of the electric motor based on the phase current of each phase detected by the current detection means;
Zero state detecting means for detecting that the d axis current or the q axis current detected by the dq axis current detecting means is in a zero state;
A specifying unit that detects a failure in the power conversion unit based on a rotation angle of the motor detected by the rotation angle detection unit when a zero state is detected by the zero state detection unit; An electric motor control system comprising:   The specifying means refers to a map that defines a correspondence relationship between the rotation angle of the motor and a failure location when the d-axis current or the q-axis current is in a zero state, and specifies the failure location in the power conversion means. The electric motor control system according to claim 1. The zero state detection unit detects the zero state when the absolute value of the d-axis current or the q-axis current detected by the dq-axis current detection unit is smaller than a predetermined determination threshold for determining the zero state. And
The specifying means includes a failure location corresponding to the map when the rotation angle of the motor detected by the rotation angle detection means is included in a predetermined angle range including the rotation angle of the motor described in the map. The motor control system according to claim 2, wherein the motor control system is specified. Prediction means for predicting the d-axis current and q-axis current of the motor when the power conversion means is not in failure based on the d-axis current command value and the q-axis current command value specified from the torque command required for the motor. Further comprising
4. The device according to claim 1, wherein the specifying unit specifies that the failure detection is not performed when the d-axis current or the q-axis current of the electric motor predicted by the prediction unit is in a zero state. 5. Motor control system.

Images