WO2019220783A1 - Failure diagnostic method, power converting device, motor module, and electric power steering device - Google Patents

Abstract

This failure diagnostic method according to an embodiment of the present disclosure diagnoses a failure of a power converting device 1000 which converts power from a power source 101 into power to be supplied to a motor 200. The failure diagnostic method includes: a step for determining whether there is a failed part among the high side and the low side of at least one H-bridge; a step for determining whether there is a failed pair among a pair of a first low-side switching element and a second high-side switching element, and a pair of a second low-side switching element and a first high-side switching element; and a step for determining whether there is a failed switching element on the basis of the determination result of whether there is a failed part and the determination result of whether there is a failed pair.

Description

Failure diagnosis method, power conversion device, motor module, and electric power steering device

The present disclosure relates to a failure diagnosis method, a power conversion device, a motor module, and an electric power steering device.

In recent years, an electromechanical integrated motor in which an electric motor (hereinafter simply referred to as “motor”), an inverter, and an ECU are integrated has been developed. Particularly in the in-vehicle field, high quality assurance is required from the viewpoint of safety. Therefore, a redundant design that can continue safe operation even when a part of the component fails is adopted. As an example of a redundant design, it is considered to provide two power conversion devices for one motor. As another example, it is considered to provide a backup microcontroller in the main microcontroller.

Patent Document 1 discloses a motor drive device having a first system and a second system. The first system is connected to the first winding set of the motor and includes a first inverter unit, a power supply relay, a reverse connection protection relay, and the like. The second system is connected to the second winding set of the motor and includes a second inverter unit, a power supply relay, a reverse connection protection relay, and the like. When there is no failure in the motor drive device, it is possible to drive the motor using both the first system and the second system. On the other hand, when a failure occurs in one of the first system and the second system, or one of the first winding group and the second winding group, the power relay is connected to the failed system or from the power source. The power supply to the system connected to the winding set is cut off. It is possible to continue motor driving using the other system that has not failed.

Patent Documents 2 and 3 also disclose a motor drive device having a first system and a second system. Even if one system or one winding set fails, motor drive can be continued by a system that does not fail.

Japan Publication: JP-A-2016-34204 Japan publication: JP-A-2016-32977 Japanese publication: JP-A-2008-132919

In the conventional technology described above, it has been required to appropriately detect a failure of a switch element included in the power conversion device.

Embodiment of this indication provides the failure diagnostic method which can diagnose appropriately the failure of the switch element with which a power converter is provided.

An exemplary failure diagnosis method of the present disclosure is a failure diagnosis method for diagnosing a failure in a power conversion device that converts electric power from a power source into electric power supplied to a motor having at least one phase winding, the electric power The conversion device includes at least one H-bridge each having a first high-side switch element, a first low-side switch element, a second high-side switch element, and a second low-side switch element, and the failure diagnosis method includes the at least Determining whether there is a failed part in the high side and low side of one H-bridge, a set of the first low side switch element and the second high side switch element, and the second low side switch element and the Determining whether there is a failed set in the set with the first high-side switch element; The first high-side switch element, the first low-side switch element, the second high-side switch element, and the second low-side are determined based on a determination result of whether there is a part and a determination result of whether there is the failed set. Determining whether there is a failed switch element among the switch elements.

An exemplary power conversion device of the present disclosure is a power conversion device that converts power from a power source into power supplied to a motor having at least one phase winding, each of the power conversion devices being a first power conversion device. At least one H bridge having a high side switch element, a first low side switch element, a second high side switch element and a second low side switch element, and a control circuit for controlling the operation of the at least one H bridge; The control circuit determines whether there is a failed part in the high side and the low side of the at least one H-bridge, the set of the first low-side switch element and the second high-side switch element, and the second Determining whether there is a failed set in the set of the low-side switch element and the first high-side switch element, The first high-side switch element, the first low-side switch element, the second high-side switch element, and the second based on the determination result of whether there is a failed part and the determination result of whether there is the failed set It is determined whether there is a failed switch element among the low-side switch elements.

According to an exemplary embodiment of the present disclosure, a failure diagnosis method capable of appropriately diagnosing a failure of a switch element included in a power conversion device, a power conversion device, a motor module including the power conversion device, and the motor module are provided. An electric power steering apparatus is provided.

FIG. 1 is a block diagram schematically illustrating a motor module according to an embodiment. FIG. 2 is a circuit diagram schematically showing the inverter unit according to the embodiment. FIG. 3A is a schematic diagram showing an A-phase H-bridge. FIG. 3B is a schematic diagram showing a B-phase H-bridge. FIG. 3C is a schematic diagram showing a C-phase H-bridge. FIG. 4 is a functional block diagram showing a controller that performs overall motor control. FIG. 5 is a diagram showing functional blocks for diagnosing the low-side and high-side faults of the A-phase H-bridge. FIG. 6 is a diagram showing functional blocks for diagnosing the low-side and high-side faults of the B-phase H-bridge. FIG. 7 is a functional block diagram for diagnosing the low-side and high-side failure of the C-phase H-bridge. FIG. 8 is a diagram illustrating functional blocks for performing phase failure diagnosis. FIG. 9 is a diagram illustrating functional blocks for performing phase failure diagnosis. FIG. 10 is a diagram illustrating a circuit model for explaining the principle of performing failure diagnosis by comparing the magnitudes of the system voltage and the voltage command value VA_ref. FIG. 11 is a diagram illustrating a modification of the functional block for performing phase failure diagnosis. FIG. 12 is a diagram showing functional blocks for performing failure diagnosis of a set of A-phase switch elements. FIG. 13 is a diagram showing functional blocks for performing failure diagnosis of a set of B-phase switch elements. FIG. 14 is a diagram showing functional blocks for performing failure diagnosis of a set of C-phase switch elements. FIG. 15 is a diagram illustrating functional blocks for performing failure diagnosis of the A-phase switch element. FIG. 16 is a diagram illustrating functional blocks for performing failure diagnosis of the B-phase switch element. FIG. 17 is a diagram illustrating functional blocks for performing failure diagnosis of the C-phase switch element. FIG. 18 is a schematic diagram showing a lookup table for determining the saturation voltage Vsat from the rotational speed ω and the current amplitude value. FIG. 19 is a graph illustrating a current waveform (sine wave) obtained by plotting the current values flowing through the windings of the A phase, B phase, and C phase of the motor when the power conversion device is controlled according to the three-phase energization control. It is. FIG. 20 is a graph illustrating a current waveform obtained by plotting the current value flowing through each of the B-phase and C-phase windings of the motor when the power conversion device is controlled in accordance with the two-phase energization control when the A-phase fails. It is. FIG. 21 is a graph exemplifying a current waveform obtained by plotting the current value flowing through each of the C-phase and A-phase windings of the motor when the B-phase has failed and the power converter is controlled according to the two-phase energization control. It is. FIG. 22 is a graph exemplifying a current waveform obtained by plotting the current value flowing through each of the A-phase and B-phase windings of the motor when the power conversion device is controlled according to the two-phase energization control when the C-phase fails. It is. FIG. 23 is a graph showing waveforms of simulation results of the actual voltage VA1 (upper side) and the actual voltage VA2 (lower side) when the low-side switch element SW_A1L has an open failure. FIG. 24 is a graph showing waveforms of simulation results of the actual voltage VB1 (upper side) and the actual voltage VB2 (lower side) when the low-side switch element SW_A1L has an open failure. FIG. 25 is a graph showing waveforms of simulation results of the actual voltage VC1 (upper side) and the actual voltage VC2 (lower side) when the low-side switch element SW_A1L has an open failure. FIG. 26 is a graph showing waveforms of simulation results of the actual voltage VA1 (upper side) and the actual voltage VA2 (lower side) when the high-side switch element SW_A1H has an open failure. FIG. 27 is a graph showing waveforms of simulation results of the actual voltage VB1 (upper side) and the actual voltage VB2 (lower side) when the high-side switch element SW_A1H has an open failure. FIG. 28 is a graph showing waveforms of simulation results of the actual voltage VC1 (upper side) and the actual voltage VC2 (lower side) when the high-side switch element SW_A1H has an open failure. FIG. 29 is a graph showing a waveform of a simulation result of the voltage command value VA_ref. FIG. 30 is a graph showing a waveform of a simulation result of the voltage command value VB_ref. FIG. 31 is a graph showing a waveform of a simulation result of the voltage command value VC_ref. FIG. 32 is a schematic diagram illustrating an electric power steering apparatus according to an exemplary embodiment.

Hereinafter, embodiments of a switch element failure diagnosis method, a power conversion device, a motor module, and an electric power steering device according to the present disclosure will be described in detail with reference to the accompanying drawings. However, in order to avoid the following description from being unnecessarily redundant and to facilitate understanding by those skilled in the art, a more detailed description than necessary may be omitted. For example, detailed descriptions of already well-known matters and repeated descriptions for substantially the same configuration may be omitted.

In the present specification, implementation of the present disclosure will be described by taking, as an example, a power conversion device that converts power from a power source into power to be supplied to a three-phase motor having three-phase (A-phase, B-phase, and C-phase) windings. A form is demonstrated. However, a power conversion device that converts electric power from a power source into electric power to be supplied to an n-phase motor having an n-phase winding (n is an integer of 4 or more) such as four-phase or five-phase, and a switch used in the device An element failure diagnosis method is also included in the scope of the present disclosure.

(Embodiment 1)
[1. Structure of motor module 2000 and power conversion apparatus 1000]
FIG. 1 schematically shows a typical block configuration of a motor module 2000 according to the present embodiment.

The motor module 2000 typically includes a power converter 1000 having the inverter unit 100 and a control circuit 300 and a motor 200. The motor module 2000 is modularized and can be manufactured and sold as, for example, an electromechanically integrated motor having a motor, a sensor, a driver, and a controller.

The power conversion apparatus 1000 can convert power from the power source 101 (see FIG. 2) into power supplied to the motor 200. The power conversion apparatus 1000 is connected to the motor 200. For example, the power conversion apparatus 1000 can convert DC power into three-phase AC power that is pseudo-sine waves of A phase, B phase, and C phase. In this specification, “connection” between components (components) mainly means electrical connection.

The motor 200 is, for example, a three-phase AC motor. The motor 200 includes an A-phase winding M1, a B-phase winding M2, and a C-phase winding M3, and is connected to the first inverter 120 and the second inverter 130 of the inverter unit 100. More specifically, the first inverter 120 is connected to one end of each phase winding of the motor 200, and the second inverter 130 is connected to the other end of each phase winding.

The control circuit 300 includes, for example, a power supply circuit 310, an angle sensor 320, an input circuit 330, a controller 340, a drive circuit 350, and a ROM 360. Each component of the control circuit 300 is mounted on, for example, one circuit board (typically a printed board). The control circuit 300 is connected to the inverter unit 100 and controls the inverter unit 100 based on input signals from the current sensor 150 and the angle sensor 320. Examples of the control method include vector control, pulse width modulation (PWM), and direct torque control (DTC). However, the angle sensor 320 may be unnecessary depending on the motor control method (for example, sensorless control).

The control circuit 300 can realize the closed loop control by controlling the target position, rotation speed, current, and the like of the motor 200. The control circuit 300 may include a torque sensor instead of the angle sensor 320. In this case, the control circuit 300 can control the target motor torque.

The power supply circuit 310 generates a power supply voltage (for example, 3V, 5V) necessary for each block in the circuit based on the voltage of the power supply 101, for example, 12V.

The angle sensor 320 is, for example, a resolver or a Hall IC. Alternatively, the angle sensor 320 is also realized by a combination of an MR sensor having a magnetoresistive (MR) element and a sensor magnet. The angle sensor 320 detects the rotation angle of the rotor (hereinafter referred to as “rotation signal”) and outputs the rotation signal to the controller 340.

The input circuit 330 receives the phase current detected by the current sensor 150 (hereinafter sometimes referred to as “actual current value”), and changes the level of the actual current value to the input level of the controller 340 as necessary. The actual current value is output to the controller 340. The input circuit 330 is, for example, an analog / digital (AD) conversion circuit.

The controller 340 is an integrated circuit that controls the entire power conversion apparatus 1000, and is, for example, a microcontroller or an FPGA (Field Programmable Gate Array). The controller 340 controls the switching operation (turn-on or turn-off) of each switch element (typically a semiconductor switch element) in the first and second inverters 120 and 130 of the inverter unit 100. The controller 340 sets the target current value according to the actual current value and the rotation signal of the rotor, generates a PWM signal, and outputs it to the drive circuit 350.

The drive circuit 350 is typically a pre-driver (sometimes called a “gate driver”). The drive circuit 350 generates a control signal (gate control signal) for controlling the switching operation of each switch element in the first and second inverters 120 and 130 of the inverter unit 100 according to the PWM signal, and supplies a control signal to the gate of each switch element. give. When the driving target is a motor that can be driven at a low voltage, the pre-driver is not necessarily required. In that case, the function of the pre-driver can be implemented in the controller 340.

The ROM 360 is, for example, a writable memory (for example, PROM), a rewritable memory (for example, flash memory), or a read-only memory. The ROM 360 stores a control program including a command group for causing the controller 340 to control the power conversion apparatus 1000. For example, the control program is temporarily expanded in a RAM (not shown) at the time of booting.

A specific circuit configuration of the inverter unit 100 will be described with reference to FIG.

FIG. 2 schematically shows a circuit configuration of the inverter unit 100 according to the present embodiment.

The power supply 101 generates a predetermined power supply voltage (for example, 12V). As the power source 101, for example, a DC power source is used. However, the power source 101 may be an AC-DC converter, a DC-DC converter, or a battery (storage battery). The power source 101 may be a single power source common to the first and second inverters 120 and 130 as shown in the figure, or may be a first power source (not shown) for the first inverter 120 and for the second inverter 130. A second power source (not shown) may be provided.

Although not shown, coils are provided between the power source 101 and the first inverter 120 and between the power source 101 and the second inverter 130. The coil functions as a noise filter, and smoothes the high frequency noise included in the voltage waveform supplied to each inverter or the high frequency noise generated by each inverter so as not to flow out to the power supply 101 side. A capacitor is connected to the power supply terminal of each inverter. The capacitor is a so-called bypass capacitor and suppresses voltage ripple. The capacitor is, for example, an electrolytic capacitor, and the capacity and the number to be used are appropriately determined according to design specifications.

The first inverter 120 has a bridge circuit composed of three legs. Each leg has a high-side switch element, a low-side switch element, and a shunt resistor. The A-phase leg includes a high-side switch element SW_A1H, a low-side switch element SW_A1L, and a first shunt resistor S_A1. The B-phase leg has a high-side switch element SW_B1H, a low-side switch element SW_B1L, and a first shunt resistor S_B1. The C-phase leg has a high-side switch element SW_C1H, a low-side switch element SW_C1L, and a first shunt resistor S_C1.

As the switch element, for example, a field effect transistor (typically MOSFET) having a parasitic diode formed therein, or a combination of an insulated gate bipolar transistor (IGBT) and a free-wheeling diode connected in parallel thereto can be used. .

The first shunt resistor S_A1 is used to detect the A-phase current IA1 flowing through the A-phase winding M1, and is connected, for example, between the low-side switch element SW_A1L and the GND line GL. The first shunt resistor S_B1 is used to detect the B-phase current IB1 flowing through the B-phase winding M2, and is connected between the low-side switch element SW_B1L and the GND line GL, for example. The first shunt resistor S_C1 is used to detect the C-phase current IC1 flowing through the C-phase winding M3, and is connected between, for example, the low-side switch element SW_C1L and the GND line GL. The three shunt resistors S_A1, S_B1, and S_C1 are connected in common with the GND line GL of the first inverter 120.

The second inverter 130 has a bridge circuit composed of three legs. Each leg has a high-side switch element, a low-side switch element, and a shunt resistor. The A-phase leg has a high-side switch element SW_A2H, a low-side switch element SW_A2L, and a shunt resistor S_A2. The B-phase leg has a high-side switch element SW_B2H, a low-side switch element SW_B2L, and a shunt resistor S_B2. The C-phase leg has a high-side switch element SW_C2H, a low-side switch element SW_C2L, and a shunt resistor S_C2.

The shunt resistor S_A2 is used to detect the A-phase current IA2, and is connected, for example, between the low-side switch element SW_A2L and the GND line GL. The shunt resistor S_B2 is used to detect the B-phase current IB2, and is connected between, for example, the low-side switch element SW_B2L and the GND line GL. The shunt resistor S_C2 is used to detect the C-phase current IC2, and is connected, for example, between the low-side switch element SW_C2L and the GND line GL. The three shunt resistors S_A2, S_B2, and S_C2 are connected in common with the GND line GL of the second inverter 130.

The current sensor 150 described above includes, for example, a shunt resistor S_A1, S_B1, S_C1, S_A2, S_B2, S_C2, and a current detection circuit (not shown) that detects a current flowing through each shunt resistor.

The A-phase leg of the first inverter 120 (specifically, a node between the high-side switch element SW_A1H and the low-side switch element SW_A1L) is connected to one end A1 of the A-phase winding M1 of the motor 200, and the second inverter The 130 A-phase leg is connected to the other end A2 of the A-phase winding M1. The B-phase leg of the first inverter 120 is connected to one end B1 of the B-phase winding M2 of the motor 200, and the B-phase leg of the second inverter 130 is connected to the other end B2 of the winding M2. The C-phase leg of the first inverter 120 is connected to one end C1 of the C-phase winding M3 of the motor 200, and the C-phase leg of the second inverter 130 is connected to the other end C2 of the winding M3.

FIG. 3A schematically shows the configuration of the A-phase H-bridge BA. FIG. 3B schematically shows the configuration of a B-phase H-bridge BB. FIG. 3C schematically shows the configuration of a C-phase H-bridge BC.

The inverter unit 100 includes A-phase, B-phase, and C-phase H-bridges BA, BB, and BC. The A-phase H bridge BA includes a high-side switch element SW_A1H and a low-side switch element SW_A1L in the leg on the first inverter 120 side, a high-side switch element SW_A2H, a low-side switch element SW_A2L in the leg on the second inverter 130 side, and a winding Has M1.

The B-phase H-bridge BB includes a high-side switch element SW_B1H and a low-side switch element SW_B1L in the leg on the first inverter 120 side, a high-side switch element SW_B2H, a low-side switch element SW_B2L in the leg on the second inverter 130 side, and a winding Has M2.

The C-phase H-bridge BC includes a high-side switch element SW_C1H and a low-side switch element SW_C1L in the leg on the first inverter 120 side, a high-side switch element SW_C2H, a low-side switch element SW_C2L in the leg on the second inverter 130 side, and a winding M3.

The control circuit 300 (specifically, the controller 340) can specify a failed switch element in the power conversion apparatus 1000 by executing a failure diagnosis described below.

For example, when the failed switch element is identified, the control circuit 300 can be switched to motor control in which a two-phase winding is energized using a two-phase H bridge other than the H bridge including the failed switch element. In this specification, energizing the three-phase winding is called “three-phase energization control”, and energizing the two-phase winding is called “two-phase energization control”. Details of the failure diagnosis will be described below.

[2. (Failure diagnosis method)
A specific example of a failure diagnosis method for diagnosing the presence or absence of a failure of the switch element of the power conversion apparatus 1000 shown in FIG. 1 will be described with reference to FIGS. The failure diagnosis method of the present disclosure can be suitably used for a power conversion device including at least one H bridge, for example, a full bridge type power conversion device. In this specification, the failure of a switch element refers to the open failure of a switch element. An open failure is a failure in which the switch element always has a high impedance. In the present specification, for example, a failure occurring in a switching element of an A-phase H-bridge may be referred to as an A-phase failure.

The outline of the failure diagnosis method of this embodiment is as follows.

In the failure diagnosis, it is determined whether there is a failed part on the high side and the low side of the H bridge BA, BB, BC. In addition, it is determined whether there is a broken set among the set of the low side switch element and the high side switch element. Based on these determination results, it is determined whether there is a failed switch element among the high-side switch element and the low-side switch element included in the power conversion device 1000.

In the failure diagnosis, for example, the current and voltage expressed in the dq coordinate system, the actual voltage indicating the voltage across the low side switch element, and the rotational speed ω of the motor are acquired. The current and voltage expressed in the dq coordinate system include a d-axis voltage Vd, a q-axis voltage Vq, a d-axis current Id, and a q-axis current Iq. In the dq coordinate system, the axis corresponding to the zero phase is represented as the z-axis. The rotation speed ω is represented by a rotation speed (rpm) at which the rotor of the motor rotates per unit time (for example, 1 minute) or a rotation speed (rps) at which the rotor rotates at unit time (for example, 1 second).

The actual voltage of the switch element will be described with reference to FIGS. 3A to 3C.

A first actual voltage and a second actual voltage are defined for each of the A-phase, B-phase, and C-phase H-bridges BA, BB, and BC. The first actual voltage indicates the voltage across the first low-side switch element in the leg on the first inverter 120 side in the H bridge of each phase. In other words, the first actual voltage corresponds to the node potential between the first high-side switch element and the first low-side switch element in the leg on the first inverter 120 side. The second actual voltage indicates the voltage across the second low-side switch element in the leg on the second inverter 130 side. In other words, the second actual voltage corresponds to the node potential between the second high-side switch element and the second low-side switch element in the leg on the second inverter 130 side. The voltage across the switch element is equal to the voltage Vds between the source and drain of the FET that is the switch element.

For the A-phase H bridge BA, the first actual voltage indicates the voltage VA1 across the low-side switch element SW_A1L shown in FIG. 3A, and the second actual voltage points across the voltage VA2 across the low-side switch element SW_A2L shown in FIG. 3A. . For the B-phase H-bridge BB, the first actual voltage indicates the voltage VB1 across the low-side switch element SW_B1L shown in FIG. 3B, and the second actual voltage indicates the voltage VB2 across the low-side switch element SW_B2L shown in FIG. 3B. . For the C-phase H-bridge BC, the first actual voltage indicates the voltage VC1 across the low-side switch element SW_C1L illustrated in FIG. 3C, and the second actual voltage indicates the voltage VC2 across the low-side switch element SW_C2L illustrated in FIG. 3C. .

Next, a failure is diagnosed based on the acquired current and voltage in the dq coordinate system, the first actual voltage, the second actual voltage, the rotation speed, and the like.

If it is determined that there is a faulty switch element, a fault signal indicating a fault of the switch element is generated and output to a motor control unit described later. For example, a failure signal is a signal that is asserted when a failure occurs.

The above-described failure diagnosis is repeatedly executed in synchronization with, for example, a period in which each phase current is measured by the current sensor 150, that is, an AD conversion period.

The algorithm for realizing the fault diagnosis method according to the present embodiment can be realized only by hardware such as an application specific integrated circuit (ASIC) or FPGA, or can be realized by a combination of a microcontroller and software. Can do. In the present embodiment, the operation subject of failure diagnosis is the controller 340 of the control circuit 300.

FIG. 4 exemplifies functional blocks of the controller 340 for performing overall motor control. FIG. 5 illustrates functional blocks for performing a low-side and high-side fault diagnosis of the A-phase H-bridge BA.

In this specification, each block in the functional block diagram is shown not in hardware units but in functional block units. The software used for motor control and failure diagnosis may be a module constituting a computer program for executing specific processing corresponding to each functional block, for example. Such a computer program is stored in the ROM 360, for example. The controller 340 can read out commands from the ROM 360 and sequentially execute each process.

The controller 340 includes, for example, a failure diagnosis unit 800 and a motor control unit 900. As described above, the failure diagnosis of the present disclosure can be suitably combined with motor control (for example, vector control), and can be incorporated into a series of processes of motor control.

Failure diagnosis unit 800 obtains d-axis current Id, q-axis current Iq, d-axis voltage Vd, q-axis voltage Vq, and rotation speed ω of motor 200 in the dq coordinate system. The fault diagnosis unit 800 further obtains the first actual voltages VA1, VB1, VC1, and the second actual voltages VA2, VB2, and VC2.

For example, the failure diagnosis unit 800 may include a pre-computation unit (not shown) that acquires Vpeak. The pre-computation unit uses the Clark transformation to convert the three-phase currents Ia, Ib and Ic obtained based on the measured value of the current sensor 150 into the currents I _α and β on the α axis in the αβ fixed coordinate system. To a current I _β of The pre-computation unit converts the currents I _α and I _β into a d-axis current Id and a q-axis current Iq in the dq coordinate system by using park conversion (dq coordinate conversion). The pre-calculation unit acquires the d-axis voltage Vd and the q-axis voltage Vq based on the currents Id and Iq, and calculates the voltage peak value Vpeak from the acquired Vd and Vq based on the following formula (1). Alternatively, the pre-computation unit can also receive Vd and Vq necessary for calculating Vpeak from the motor control unit 900 that performs vector control. For example, the pre-computation unit acquires Vpeak in synchronization with the period in which each phase current is measured by the current sensor 150.
Vpeak = (2/3) ^1/2 (Vd ² + Vq ² ) ^1/2 formula (1)

Failure diagnosis unit 800 refers to lookup table 940 (FIG. 18) and determines saturation voltage Vsat based on currents Id and Iq and rotational speed ω.

FIG. 18 schematically shows a look-up table (LUT) 940 for determining the saturation voltage Vsat from the rotation speed ω and the current amplitude value. The LUT 940 associates the relationship between the saturation voltage Vsat and the input of the current amplitude value (Id ² + Iq ² ) ^1/2 determined based on the d-axis current and the q-axis current and the rotational speed ω of the motor 200.

The rotation speed ω is calculated based on, for example, a rotation signal from the angle sensor 320. Alternatively, the rotational speed ω can be estimated using, for example, a known sensorless control method. The actual voltage of each switch element is measured by a drive circuit (predriver) 350, for example.

Table 1 illustrates the configuration of the LUT 940 that can be used for failure diagnosis. In motor control, Id is generally treated as zero. Therefore, the current amplitude value is equal to Iq. Table 1 lists Iq (A). The saturation voltage Vsat is determined from the acquired current amplitude value Iq and the rotational speed ω. Alternatively, for example, a value set in advance before driving may be used as the saturation voltage Vsat. For example, a constant value (for example, about 0.1 V) depending on the system may be used as the saturation voltage Vsat.

Failure diagnosis unit 800 obtains current peak command value Ipeak_ref and voltage peak value Vpeak expressed in the dq coordinate system. In addition, voltage command values VA_ref, VB_ref and VC_ref are obtained.

The failure diagnosis unit 800 diagnoses the presence or absence of a switch element failure based on the actual voltage, voltage peak value, saturation voltage, current peak command value, and voltage command value described above.

The failure diagnosis unit 800 generates a failure signal indicating the failed switch element based on the diagnosis result, and outputs the failure signal to the motor control unit 900.

The motor control unit 900 generates a PWM signal that controls the overall switching operation of the switch elements of the first and second inverters 120 and 130 using, for example, vector control. The motor control unit 900 outputs a PWM signal to the drive circuit 350. For example, when a failure signal is asserted, the motor control unit 900 can switch the motor control from the three-phase energization control to the two-phase energization control.

In this specification, for convenience of explanation, each functional block may be expressed as a unit. Naturally, these notations are not used with the intention of restricting each functional block to hardware or software.

When each functional block is implemented as software in the controller 340, the execution subject of the software may be the core of the controller 340, for example. As described above, the controller 340 can be implemented by an FPGA. In that case, all or some of the functional blocks may be realized by hardware.

演算 By distributing processing using multiple FPGAs, the computing load of a specific computer can be distributed. In that case, all or some of the functional blocks shown in FIG. 4 to FIG. 18 may be distributed and implemented in a plurality of FPGAs. The plurality of FPGAs can be connected to each other by, for example, an in-vehicle control area network (CAN), and can transmit and receive data.

(H bridge high side and low side failure diagnosis)
FIG. 5 shows a failure diagnosis unit 810A for diagnosing the presence or absence of a low-side and high-side failure of the A-phase H bridge BA. FIG. 6 shows a failure diagnosis unit 810B for diagnosing the presence or absence of a low-side and high-side failure of the B-phase H-bridge BB. FIG. 7 shows a failure diagnosis unit 810C for diagnosing the presence / absence of a low-side and high-side failure of the C-phase H-bridge BC. The fault diagnosis units 810A, 810B, and 810C have substantially the same functional blocks, but input actual voltages are different from each other. The failure diagnosis unit 800 includes failure diagnosis units 810A, 810B, and 810C shown in FIGS.

Each of the fault diagnosis units 810A, 810B, and 810C includes multipliers 811 and 812, adders 813, 814, and 815, comparators 816 and 817, and a logic circuit OR818. Each of failure diagnosis units 810A, 810B, and 810C further includes multipliers 821 and 822, adders 823, 824, and 825, comparators 826 and 827, and a logic circuit OR 828.

First, the diagnosis process for the presence or absence of a failure on the low side and the high side of the A-phase H bridge BA will be described with reference to FIG.

The multiplier 811 multiplies the voltage peak value Vpeak by a constant “−1/2”. The multiplier 812 multiplies the saturation voltage Vsat by a constant “−1”. Adder 813 adds the output values of multipliers 811 and 812. The adder 814 adds the actual voltage VA1 and the output value of the adder 813 to calculate a failure diagnosis voltage VA1L_FD represented by the following formula (2).
VA1L_FD = VA1-[(Vpeak / 2) + Vsat] Formula (2)

The comparator 816 compares “VA1L_FD” with “zero”. When VA1L_FD is equal to or smaller than zero (VA1L_FD ≦ 0), the comparator 816 outputs “0” indicating that the actual voltage VA1 is normal to the logic circuit OR818. When VA1L_FD is larger than zero (VA1L_FD> 0), the comparator 816 outputs “1” indicating that the actual voltage VA1 is abnormal to the logic circuit OR818.

Similarly, the adder 815 adds the actual voltage VA2 and the output value of the adder 813 to calculate the failure diagnosis voltage VA2L_FD.

The comparator 817 compares “VA2L_FD” with “zero”. When VA2L_FD is equal to or less than zero (VA2L_FD ≦ 0), the comparator 817 outputs “0” indicating that the actual voltage VA2 is normal to the logic circuit OR818. When VA2L_FD is larger than zero (VA2L_FD> 0), the comparator 817 outputs “1” indicating that the actual voltage VA2 is abnormal to the logic circuit OR818.

The logic circuit OR 818 takes the logical sum of the output signals of the comparators 816 and 817. The logic circuit OR 818 outputs a failure signal AL_FD indicating whether or not there is a failure on the low side of the A-phase H bridge BA.

When the output signals of the comparators 816 and 817 are all “0”, the logic circuit OR 818 outputs “0” indicating normality as the failure signal AL_FD. When at least one of the output signals of the comparators 816 and 817 is “1”, the logic circuit OR 818 outputs “1” indicating a failure as the failure signal AL_FD.

For example, when an open failure occurs in the low-side switch element SW_A2L, current hardly flows even if the switch element SW_A2L is in an ON state. Since almost no current flows through the switch element SW_A2L, an extra voltage is applied to the switch element SW_A1L. The actual voltage VA1 increases and VA1_FD> 0.

The multiplier 821 multiplies the voltage peak value Vpeak by a constant “½”. The multiplier 822 multiplies the saturation voltage Vsat by a constant “1”. Adder 823 adds the output values of multipliers 821 and 822. The adder 824 adds the actual voltage VA1 and the output value of the adder 823 to calculate a failure diagnosis voltage VA1H_FD represented by the following formula (3).
VA1H_FD = VA1 + [(Vpeak / 2) + Vsat] Formula (3)

The comparator 826 compares “VA1H_FD” with “zero”. When VA1H_FD is zero or more (VA1H_FD ≧ 0), the comparator 826 outputs “0” indicating that the actual voltage VA1 is normal to the logic circuit OR828. When VA1H_FD is less than zero (VA1H_FD <0), the comparator 826 outputs “1” indicating that the actual voltage VA1 is abnormal to the logic circuit OR826.

Similarly, the adder 825 adds the actual voltage VA2 and the output value of the adder 823 to calculate the failure diagnosis voltage VA2H_FD.

The comparator 827 compares “VA2H_FD” with “zero”. When VA2H_FD is zero or more (VA2H_FD ≧ 0), the comparator 827 outputs “0” indicating that the actual voltage VA2 is normal to the logic circuit OR828. When VA2H_FD is less than zero (VA2H_FD <0), the comparator 827 outputs “1” indicating that the actual voltage VA2 is abnormal to the logic circuit OR828.

The logic circuit OR 828 takes the logical sum of the output signals of the comparators 826 and 827. The logic circuit OR828 outputs a failure signal AH_FD indicating whether or not there is a failure on the high side of the A-phase H bridge BA.

When the output signals of the comparators 826 and 827 are all “0”, the logic circuit OR 828 outputs “0” indicating normality as the failure signal AH_FD. When at least one of the output signals of the comparators 826 and 827 is “1”, the logic circuit OR 828 outputs “1” indicating a failure as the failure signal AH_FD.

The failure diagnosis unit 810B shown in FIG. 6 performs the same processing as the failure diagnosis unit 810A, and diagnoses the presence or absence of a low-side and high-side failure of the B-phase H-bridge BB. Instead of the actual voltages VA1 and VA2, actual voltages VB1 and VB2 are input to the failure diagnosis unit 810B. The logic circuit OR 818 outputs a failure signal BL_FD indicating the presence or absence of a low-side failure of the B-phase H bridge BB. The logic circuit OR828 outputs a failure signal BH_FD indicating whether or not the high-side failure of the B-phase H-bridge BB exists. Since the other processing of the failure diagnosis unit 810B is the same as that of the failure diagnosis unit 810A, detailed description is omitted here.

The failure diagnosis unit 810C shown in FIG. 7 executes the same processing as the failure diagnosis unit 810A, and diagnoses the presence or absence of a low-side and high-side failure of the C-phase H-bridge BC. Instead of the actual voltages VA1 and VA2, actual voltages VC1 and VC2 are input to the failure diagnosis unit 810C. The logic circuit OR 818 outputs a failure signal CL_FD indicating whether or not there is a failure on the low side of the C-phase H-bridge BC. The logic circuit OR 828 outputs a failure signal CH_FD indicating the presence or absence of a high-side failure of the C-phase H bridge BC. Since the other processing of the failure diagnosis unit 810C is the same as that of the failure diagnosis unit 810A, detailed description is omitted here.

(Phase failure diagnosis)
The failure diagnosis unit 800 has a failure diagnosis unit 830 for diagnosing the phase failure shown in FIG. The failure diagnosis unit 830 diagnoses a failure in the A phase, the B phase, and the C phase using the voltage peak value Vpeak, the current peak command value Ipeak_ref, and the voltage command values VA_ref, VB_ref, and VC_ref.

FIG. 9 is a block diagram illustrating a failure diagnosis unit 830 that diagnoses a failure in the A phase, the B phase, and the C phase. The failure diagnosis unit 830 includes, for example, a gain unit 831, a limit determination unit 832, an LPF (low pass filter) 833, an adder 834, a multiplier 835, absolute value calculators 836A, 836B, 836C, adders 837A, 837B, 837C, It has signal generation units 838A, 838B and 838C.

The gain unit 831 multiplies the current peak command value Ipeak_ref by the gain R. The current peak command value Ipeak_ref indicates the peak value of the current amplitude in the dq coordinate system, and is specifically calculated based on the following formula (4). Here, Idref represents a d-axis current command value on the d-axis, Iqref represents a q-axis current command value on the q-axis, and Izref represents a zero-phase current command value. abs (X) represents the absolute value of X. The gain R represents the electrical characteristics of the entire circuit system including the H bridge. For example, the gain R is determined in consideration of the influence of the dead time of the switch element, and corresponds to the resistance [Ω] of the entire circuit.
Ipeak_ref = (2/3) ^1/2 (Idref ² + Iqref ² ) ^1/2 + abs (Izref) / 3 ^1/2 equation (4)

For example, the core of the controller 340 determines the current command values Idref, Iqref, and Izref based on the rotational speed and the speed command value detected by the angle sensor 320 and outputs them to the failure diagnosis unit 800. For example, a pre-computation unit (not shown) calculates Ipeak_ref based on the current command values Idref, Iqref, and Izref, and outputs it to the gain unit 831. The gain unit 831 outputs Ipeak_ref · R to the limit determination unit 832.

The limit determination unit 832 determines whether or not the product of the current peak command value Ipeak_ref and the gain R is within an allowable range. That is, the limit determination unit 832 determines whether Ipeak_ref · R is within the allowable range. The allowable range means the upper limit value of the input voltage during normal operation.

It is preferable that the voltage peak value Vpeak is low-pass filtered by the general-purpose LPF 833 before the subsequent operation of adding the voltage peak value Vpeak is executed. As a result, the voltage peak value Vpeak expressed only by the fundamental wave can be acquired.

The adder 834 adds the output (Ipeak_ref · R) from the limit determination unit 832 and the output Vpeak from the LPF 833. The adder 834 outputs Ipeak_ref · R + Vpeak to the multiplier 835.

The multiplier 835 multiplies the output Ipeak_ref · R + Vpeak from the adder 834 by “−1”. In this specification, the output voltage (Ipeak_ref · R + Vpeak) from the adder 834 is referred to as “system voltage”. Multiplier 835 outputs the system voltage to A-phase adder 837A, B-phase adder 837B, and C-phase adder 837C.

Hereinafter, the fault diagnosis process of the H bridge will be described by taking the fault diagnosis of the A phase as an example.

The failure diagnosis unit 830 diagnoses the failure of the A-phase H-bridge BA based on the comparison result between the system voltage and the magnitude of the A-phase voltage command value VA_ref. If the magnitude of voltage command value VA_ref is greater than the system voltage, failure diagnosis unit 800 determines that A-phase H-bridge BA has failed. If the magnitude of voltage command value VA_ref is equal to or lower than the system voltage, failure diagnosis unit 800 determines that A-phase H-bridge BA has not failed. In this embodiment, the multiplier 835, the absolute value calculator 836A, and the adder 837A are used to compare the system voltage with the magnitude of the A-phase voltage command value VA_ref.

Adder 837A adds the output voltage from multiplier 835 and the magnitude of A-phase voltage command value VA_ref. The reason why the absolute value calculator 836A takes the absolute value of the voltage command value VA_ref is that the open failure of both the high-side switch element and the low-side switch element of the H-bridge is a target of failure diagnosis.

Referring to FIGS. 3A to 3C again.

The A-phase voltage command value VA_ref includes the voltage command value VA1_ref for the high-side switch element SW_A1H or the low-side switch element SW_A1L (node between both switch elements) of the H-bridge BA, the high-side switch element SW_A2H, or the low-side switch element SW_A2L. This is given by the difference from the voltage command value VA2_ref for (a node between both switch elements). The B phase voltage command value VB_ref and the C phase voltage command value VC_ref are also given in the same manner as the A phase voltage command value VA_ref. Voltage command values VA_ref, VB_ref, and VC_ref are calculated based on Expression (5).
VA_ref = VA1_ref−VA2_ref
VB_ref = VB1_ref−VB2_ref Equation (5)
VC_ref = VC1_ref−VC2_ref

When the addition value from the adder 837A is larger than zero, the signal generation unit 838A determines that the H-bridge BA has failed, generates a failure signal A_FD, and outputs it to the motor control unit 900. For example, the failure signal A_FD is assigned to a 1-bit signal, and the level of the failure signal A_FD at the normal time is set to the low level “0”. When the signal generation unit 838A detects a failure of the H bridge BA, the signal generation unit 838A generates a failure signal A_FD having a high level “1”. In other words, the signal generation unit 838A asserts the failure signal A_FD.

The failure diagnosis unit 830 generates B-phase and C-phase failure signals B_FD and C_FD in the same manner as the A phase, and outputs them to the motor control unit 900.

FIG. 10 shows a circuit model for explaining the principle of performing a fault diagnosis of the H-bridge by comparing the magnitudes of the system voltage and the voltage command value VA_ref.

As described above, the gain R corresponds to the resistance [Ω] of the entire circuit. In the circuit model shown in FIG. 10, a current command value Ipeak_ref is a current flowing through the entire circuit, a gain R is an internal resistance of the circuit, and Vpeak is an input voltage. Failure diagnosis unit 800 diagnoses a failure of H-bridge BA based on voltage command value VA_ref, input voltage Vpeak, and voltage drop (Ipeak_ref · R) of internal resistance R.

When the H-bridge BA is not faulty, the difference (Vpeak + Ipeak_ref · R−VA_ref) between Vpeak + Ipeak_ref · R and the voltage command value VA_ref is zero or less. That is, the relationship Vpeak + Ipeak_ref · R≈VA_ref is established.

When the H bridge BA fails, the voltage command value VA_ref increases. For this reason, the difference (Vpeak + Ipeak_ref · R−VA_ref) does not become zero, and Vpeak + Ipeak_ref · R <VA_ref, and the balance is lost.

FIG. 11 shows a modified example of the functional block for performing the H bridge failure diagnosis.

In this modification, subtracters 839A, 839B and 839C are used instead of the multiplier 835 and adders 837A, 837B and 837C. Failure diagnosis unit 830 determines that H-bridge BA has failed when the difference value obtained by subtracting the system voltage from the magnitude of A-phase voltage command value VA_ref is greater than zero. The failure diagnosis unit 830 determines that the H-bridge BA has not failed when the difference value is zero or less. The failure diagnosis unit 800 can determine the B phase and the C phase similarly to the A phase.

A lookup table (LUT) 801 can be used in place of the gain unit 831 and the limit determination unit 832. The LUT 801 is a table that associates the relationship between the current peak command value Ipeak_ref and the speed input indicating the rotation speed of the motor 200 and the output voltage Vsat. The failure diagnosis unit 830 refers to the LUT 801 and determines the output voltage Vsat based on the acquired current peak command value Ipeak_ref and the rotation speed speed. The failure diagnosis unit 830 may acquire the system voltage by performing an operation of adding the voltage peak value Vpeak to the determined output voltage Vsat.

(Failure diagnosis of switch element set)
Next, failure diagnosis of a set of a low side switch element and a high side switch element will be described.

FIG. 12 shows a failure diagnosis unit 850A for diagnosing the presence / absence of a failure in the combination of the low-side switch element and the high-side switch element of the A-phase H bridge BA. FIG. 13 shows a failure diagnosis unit 850B for diagnosing the presence / absence of a failure in a set of a low-side switch element and a high-side switch element of a B-phase H-bridge BB. FIG. 14 shows a failure diagnosis unit 850C for diagnosing the presence / absence of a failure in a set of a low-side switch element and a high-side switch element of a C-phase H-bridge BC. Fault diagnosis units 850A, 850B, and 850C have substantially the same functional blocks, but input fault signals and voltage command values are different from each other. The failure diagnosis unit 800 includes failure diagnosis units 850A, 850B, and 850C shown in FIGS.

Each of the failure diagnosis units 850A, 850B, and 850C includes a sign function unit 851, a multiplier 852, and comparators 853 and 854.

First, referring to FIG. 12, a diagnosis process for the presence or absence of a failure of a set of the low-side switch element and the high-side switch element of the A-phase H-bridge BA will be described.

The sign function unit 851 outputs “1”, “−1”, or “0” according to the voltage command value VA_ref. In this example, when the voltage command value VA_ref is a positive value, “1” is output. When the voltage command value VA_ref is a negative value, “−1” is output. When the voltage command value VA_ref is zero, “0” is output.

The multiplier 852 multiplies the fault signal A_FD indicating the presence or absence of the A-phase fault by the output value of the sign function unit 851 and outputs the result.

The comparator 853 compares the output value of the multiplier 852 with “zero”. The comparator 853 outputs a failure signal A1L2H_FD indicating whether or not there is a failure in the pair of the low side switch element SW_A1L and the high side switch element SW_A2H.

When the output value of the multiplier 852 is less than or equal to zero (output value ≦ 0), the comparator 853 outputs “0” indicating that the pair of the low-side switch element SW_A1L and the high-side switch element SW_A2H is normal. . When the output value of the multiplier 852 is greater than zero (output value> 0), the comparator 853 outputs “1” indicating that the pair of the low-side switch element SW_A1L and the high-side switch element SW_A2H is abnormal.

As shown in the above equation (5), the voltage command value VA_ref is a difference between the voltage command value VA1_ref and the voltage command value VA2_ref. Referring to FIG. 3A, when an open failure occurs in at least one of switch elements SW_A1L and SW_A2H, it becomes difficult for a current to flow through a current path including switch element SW_A2H, winding M1, and switch element SW_A1L. The voltage command value VA1_ref increases as a current flows through the current path. As a result, the voltage command value VA_ref becomes a value larger than zero.

When the fault signal A_FD indicates “1” and the output value of the sign function unit 851 indicates “1”, the output value of the multiplier 852 is greater than zero. Thereby, it can be determined that the pair of the low-side switch element SW_A1L and the high-side switch element SW_A2H is abnormal.

The comparator 854 compares the output value of the multiplier 852 with “zero”. The comparator 854 outputs a failure signal A2L1H_FD indicating whether or not there is a failure of the pair of the low side switch element SW_A2L and the high side switch element SW_A1H.

When the output value of the multiplier 852 is zero or more (output value ≧ 0), the comparator 854 outputs “0” indicating that the pair of the low-side switch element SW_A2L and the high-side switch element SW_A1H is normal. . When the output value of the multiplier 852 is smaller than zero (output value <0), the comparator 854 outputs “1” indicating that the pair of the low side switch element SW_A2L and the high side switch element SW_A1H is abnormal.

Referring to FIG. 3A, when an open failure occurs in at least one of switch elements SW_A2L and SW_A1H, it is difficult for a current to flow through a current path including switch element SW_A1H, winding M1, and switch element SW_A2L. The voltage command value VA2_ref increases as a current flows through this current path. As a result, the voltage command value VA_ref becomes a value smaller than zero.

When the fault signal A_FD indicates “1” and the output value of the sign function unit 851 indicates “−1”, the output value of the multiplier 852 becomes smaller than zero. Thereby, it can be determined that the pair of the low-side switch element SW_A2L and the high-side switch element SW_A1H is abnormal.

The failure diagnosis unit 850B shown in FIG. 13 executes the same processing as the failure diagnosis unit 850A, and diagnoses the presence / absence of a failure in the pair of the low-side switch element and the high-side switch element of the B-phase H-bridge BB. Failure diagnosis unit 850B receives failure signal B_FD and voltage command value VB_ref instead of failure signal A_FD and voltage command value VA_ref. The comparator 853 outputs a failure signal B1L2H_FD indicating whether or not there is a failure of the pair of the low side switch element SW_B1L and the high side switch element SW_B2H. The comparator 854 outputs a failure signal B2L1H_FD indicating whether or not there is a failure of the pair of the low side switch element SW_B2L and the high side switch element SW_B1H. Since other processes of the failure diagnosis unit 850B are the same as those of the failure diagnosis unit 850A, detailed description thereof is omitted here.

The failure diagnosis unit 850C shown in FIG. 14 executes the same processing as that of the failure diagnosis unit 850A, and diagnoses the presence / absence of a failure in the combination of the low-side switch element and the high-side switch element of the C-phase H-bridge BC. Fault diagnosis unit 850C receives fault signal C_FD and voltage command value VC_ref instead of fault signal A_FD and voltage command value VA_ref. The comparator 853 outputs a failure signal C1L2H_FD indicating whether or not there is a failure of the pair of the low side switch element SW_C1L and the high side switch element SW_C2H. The comparator 854 outputs a failure signal C2L1H_FD indicating whether or not there is a failure of the pair of the low side switch element SW_C2L and the high side switch element SW_C1H. Since the other processing of the failure diagnosis unit 850C is the same as that of the failure diagnosis unit 850A, detailed description thereof is omitted here.

(Identification of failed switch element)
Next, a process for determining whether there is a failed switch element among the high-side switch element and the low-side switch element included in the power conversion device 1000 will be described.

FIG. 15 shows a failure diagnosis unit 870A for diagnosing the presence or absence of a failure of the switch element of the A-phase H bridge BA. FIG. 16 shows a failure diagnosis unit 870B for diagnosing the presence or absence of a failure of a switch element included in the B-phase H bridge BB. FIG. 17 shows a failure diagnosis unit 870C for diagnosing the presence or absence of a failure of a switch element included in the C-phase H-bridge BC.

Failure diagnosis units 870A, 870B, and 870C have substantially the same functional blocks, but input signals are different from each other. The failure diagnosis unit 800 includes failure diagnosis units 870A, 870B, and 870C shown in FIGS. Each of failure diagnosis units 870A, 870B, and 870C includes logic circuits AND871, 872, 873, and 874.

First, a diagnosis process for determining whether or not a switch element included in the A-phase H-bridge BA has a failure will be described.

The failure signal AH_FD and the failure signal A2L1H_FD are input to the logic circuit AND871. The failure signal AH_FD indicates whether or not there is a failure on the high side of the A-phase H bridge BA. The failure signal A2L1H_FD indicates the presence / absence of a failure of the pair of the low side switch element SW_A2L and the high side switch element SW_A1H.

The logic circuit AND871 outputs a failure signal A1H_FD indicating whether or not the A-phase high-side switch element SW_A1H has failed to the motor control unit 900.

When both the failure signal AH_FD and the failure signal A2L1H_FD are “1” indicating the failure, the logic circuit AND871 outputs “1”. The failure signal A1H_FD being “1” indicates that the high-side switch element SW_A1H has failed. When at least one of the failure signal AH_FD and the failure signal A2L1H_FD is “0” indicating normality, the logic circuit AND871 outputs “0”. The failure signal A1H_FD being “0” indicates that the high-side switch element SW_A1H is normal.

The fact that the high side of the A-phase H bridge BA has failed and the combination of the low side switch element SW_A2L and the high side switch element SW_A1H has failed means that the high side switch element SW_A1H has failed. Represents. When the failure signal A1H_FD output from the logic circuit AND871 is “1”, it can be specified that the high-side switch element SW_A1H has failed.

Similarly, the failure signal AH_FD and the failure signal A1L2H_FD are input to the logic circuit AND872. The logic circuit AND872 outputs a failure signal A2H_FD indicating whether or not the A-phase high-side switch element SW_A2H has failed to the motor control unit 900.

The failure signal AL_FD and the failure signal A1L2H_FD are input to the logic circuit AND873. The logic circuit AND873 outputs a failure signal A1L_FD indicating whether or not the A-phase low-side switch element SW_A1L has failed to the motor control unit 900.

The failure signal AL_FD and the failure signal A2L1H_FD are input to the logic circuit AND874. The logic circuit AND874 outputs a failure signal A2L_FD indicating whether or not the A-phase low-side switch element SW_A2L has failed to the motor control unit 900.

Since the process executed by the failure diagnosis unit 870B shown in FIG. 16 is the same as that of the failure diagnosis unit 870A, detailed description thereof is omitted here. The failure diagnosis unit 870B executes the same process as the failure diagnosis unit 870A, and executes a diagnosis process of whether or not there is a failure in the switch element of the B-phase H bridge BB. The failure diagnosis unit 870B outputs failure signals B1H_FD, B2H_FD, B1L_FD, and B2L_FD that indicate the presence or absence of failure of the switch elements SW_B1H, SW_B2H, SW_B1L, and SW_B2L to the motor control unit 900. When a failure occurs in the B-phase H-bridge BB, it is possible to identify which switch element has failed.

Since the process executed by the failure diagnosis unit 870C shown in FIG. 17 is the same as that of the failure diagnosis unit 870A, detailed description thereof is omitted here. The failure diagnosis unit 870C executes the same process as the failure diagnosis unit 870A, and executes a diagnosis process of whether or not there is a failure of the switch element included in the C-phase H-bridge BC. The failure diagnosis unit 870C outputs failure signals C1H_FD, C2H_FD, C1L_FD, and C2L_FD that indicate the presence / absence of failure of the switch elements SW_C1H, SW_C2H, SW_C1L, and SW_C2L to the motor control unit 900. When a failure has occurred in the C-phase H-bridge BC, it is possible to identify which switch element has failed.

The motor control unit 900 changes the motor control according to the failure signal output by the failure diagnosis unit 800. For example, the motor control is switched from three-phase energization control to two-phase energization control. For example, when a failed switch element is specified, two-phase energization control using the remaining two phases other than the phase including the failed switch element is performed. For example, when it is determined that a switch element of the A-phase H bridge BA has failed, the motor control unit 900 turns off all the switch elements of the A-phase H bridge BA. Then, two-phase energization control using the remaining B-phase and C-phase H-bridges BB and BC is performed. Therefore, even if one of the three phases fails, the power conversion apparatus 1000 can continue to drive the motor.

FIG. 19 exemplifies a current waveform (sine wave) obtained by plotting the current values flowing in the A-phase, B-phase, and C-phase windings of the motor 200 when the power conversion apparatus 1000 is controlled according to the three-phase energization control. doing. FIG. 20 is obtained by plotting the values of current flowing through the B-phase and C-phase windings of the motor 200 when the power conversion apparatus 1000 is controlled according to the two-phase energization control when the A-phase H-bridge BA fails. The current waveform is illustrated. The horizontal axis represents the motor electrical angle (deg), and the vertical axis represents the current value (A). In the current waveforms of FIGS. 19 and 20, current values are plotted for every electrical angle of 30 °. _Ipk represents the maximum current value (peak current value) of each phase.

For reference, when the B-phase H-bridge BB fails, FIG. 21 plots the values of current flowing through the A-phase and C-phase windings of the motor 200 when the power converter 1000 is controlled according to the two-phase energization control. The current waveform obtained in this way is illustrated. In FIG. 22, when the C-phase H-bridge BC fails, the current values flowing in the A-phase and B-phase windings of the motor 200 are plotted when the power converter 1000 is controlled according to the two-phase energization control. The current waveform is illustrated.

In the present embodiment, the order of processing between the above-described failure diagnosis units 810A, 810B, 810C, 830, 850A, 850B, and 850C is arbitrary. For example, after determining whether there is a set of failed switch elements, it may be determined whether there is a failed part in the high side and low side of the H-bridge, or vice versa. These determinations may be processed in parallel.

Further, only a part of the processing may be executed without executing all the processing of the failure diagnosis units 810A, 810B, 810C, 830, 850A, 850B, and 850C. For example, in the process of determining whether there is a faulty part, if it is determined that there is a faulty part before determining the presence / absence of faults in all the high-side and low-side parts of the H-bridge BA, BB, BC, It is not necessary to determine whether there is a failure in the part. For example, if it is determined that the high side of the H-bridge BA is out of order before determining the presence / absence of the failure of all the parts, it is not necessary to determine the presence / absence of the failure of the remaining parts. If it is determined that the high side of the H-bridge BA has failed, it is possible to identify the failed switch element even if the processing related to the remaining high side and low side is omitted.

Also, for example, in the process of determining whether there is a set of failed switch elements, if it is determined that there is a failed set before determining the presence / absence of all sets of faults, the determination of the presence / absence of faults in the remaining sets is It does not have to be done. For example, if it is determined that the combination of the low-side switch element SW_A1L and the high-side switch element SW_A2H has failed before determining the presence / absence of failure in all the pairs, the presence / absence of failure in the remaining pairs is not determined. May be. When it is determined that the set of the low-side switch element SW_A1L and the high-side switch element SW_A2H is out of order, the failed switch element can be identified even if the processing related to the remaining set is omitted.

Further, when it is determined that there is a failure in the step of determining whether there is a failed part, it may be determined whether there is a failed set only for a set of 2n-1 switch elements of n H bridges. . For example, if it is determined that 5 out of 6 sets included in the three-phase H-bridges BA, BB, and BC are not in failure, the remaining 1 set does not perform failure diagnosis, and the remaining 1 It can be identified that the set is out of order.

Further, when it is determined that there is a failure in the step of determining whether there is a set of failed switch elements, only the 2n-1 of the 2n high side and low side of the n H bridges have failed. It may be determined whether there is. For example, if it is determined that five of the six parts included in the three-phase H-bridges BA, BB, and BC do not fail, the remaining one part does not need to be diagnosed. It can be determined that the remaining one part is out of order.

Further, the processing of the failure diagnosis units 870A, 870B, and 870C may be performed only for the phase determined to be in failure.

Thus, the amount of calculation can be reduced by omitting some of the plurality of processes. By reducing the amount of calculation, when a failure occurs, the failure can be dealt with in a shorter time.

Hereinafter, the validity of the algorithm used for the failure diagnosis according to the present disclosure will be shown using the “Rapid Control Prototyping (RCP) System” of dSPACE and the Matlab / Simlink of MathWorks. For this verification, a model of a surface magnet type (SPM) motor used in an electric power steering (EPS) apparatus, which is controlled by vector control, was used. In the verification, the q-axis current command value Iq_ref was set to 3A, and the d-axis current command value Id_ref and the zero-phase current command value Iz_ref were set to 0A. The rotation speed ω of the motor was set to 1200 rpm. In the simulation, an open failure occurs in the low-side switch element SW_A1L of the first inverter 120 at time 1.641s.

23 to 31 show the simulation results of the waveforms of the signals. The vertical axis of each graph represents voltage (V), and the horizontal axis represents time (s).

FIG. 23 shows waveforms of the actual voltage VA1 (upper side) and the actual voltage VA2 (lower side) when the low-side switch element SW_A1L has an open failure. FIG. 24 shows waveforms of the actual voltage VB1 (upper side) and the actual voltage VB2 (lower side) when the low-side switch element SW_A1L has an open failure. FIG. 25 shows waveforms of the actual voltage VC1 (upper side) and the actual voltage VC2 (lower side) when the low-side switch element SW_A1L has an open failure.

After the low-side switch element SW_A1L has an open failure at time 1.641s, it can be seen that the lower peak value of the actual voltage VA1 increases as shown in FIG. It can also be seen that the upper peak value of the actual voltage VA2 is increasing. That is, the absolute value of the upper peak value of the actual voltage VA2 increases. As shown in FIGS. 24 and 25, the actual voltages VB1, VB2, VC1, and VC2 have a small degree of change.

Even in normal operation, it may occur that the actual voltage becomes slightly higher than Vpeak / 2. However, in this embodiment, the value obtained by adding the saturation voltage Vsat to Vpeak / 2 is compared with the actual voltage. For this reason, it can be determined that a failure has occurred only when an actual voltage that has changed significantly, such as the actual voltage VA2 shown in FIG. When the actual voltage is higher than Vpeak / 2 in the normal operation, the failure determination accuracy can be improved by not determining the failure.

FIG. 26 shows waveforms of the first actual voltage VA1 (upper side) and the second actual voltage VA2 (lower side) of the A phase when the high side switch element SW_A1H in the A phase H bridge BA has an open failure. FIG. 27 shows waveforms of the first actual voltage VB1 (upper side) and the second actual voltage VB2 (lower side) of the B phase when the high-side switch element SW_A1H has an open failure. FIG. 28 shows waveforms of the C-phase first actual voltage VC1 (upper side) and the second actual voltage VC2 (lower side) when the high-side switch element SW_A1H has an open failure.

After the open failure of the high-side switch element SW_A1H of the A-phase H-bridge BA at time 1.543s, it can be seen that the upper peak value of the first actual voltage VA1 is lowered as shown in FIG. It can also be seen that the lower peak value of the second actual voltage VA2 is decreasing (the absolute value of the lower peak value is increased). As shown in FIGS. 27 and 28, the first actual voltages VB1 and VC1 and the second actual voltages VB2 and VC2 have a small degree of change.

FIG. 29 shows a waveform of the voltage command value VA_ref. FIG. 30 shows a waveform of the voltage command value VB_ref. FIG. 31 shows a waveform of the voltage command value VC_ref. The vertical axis represents voltage (V).

It can be seen that the voltage command value VA_ref rises as shown in FIG. 29 after the high-side switch element SW_A2H of the A-phase H-bridge BA at time 1.543s has an open failure. As shown in FIGS. 30 and 31, the voltage command values VB_ref and VC_ref are not increased.

In the above embodiment, it is possible to detect a failure of the switch element even in the process during the failure diagnosis. However, the failure of the switch element is not determined at some point in the middle. In this embodiment, based on the determination result of whether there is a faulty part in the high side and the low side of the H bridge and the determination result of whether there is a set of faulty switch elements, the faulty switch element is determined. Do. Thereby, the accuracy of failure diagnosis can be improved.

According to this embodiment, it is possible to identify a switch element that has an open failure among the switch elements of the H bridge. The failure diagnosis of the present disclosure can be realized by a simple algorithm. For this reason, for example, an advantage of reducing the circuit size or the memory size can be obtained in mounting 340 to the controller.

The failure diagnosis method of the present disclosure can be suitably used for a full bridge type power conversion device. The full bridge includes a one-phase H-bridge structure, for example, the circuit structure shown in FIG. 3A. By utilizing the above-described failure diagnosis method for full bridge failure diagnosis, a full bridge failure can be detected.

For example, the full-bridge type power converter controls the switching operation of the H-bridge BA having the high-side switch element SW_A1H, the high-side switch element SW_A2H, the low-side switch element SW_A1L, and the low-side switch element SW_A2L, and the switch element of the H-bridge BA. And a control circuit 300. The control circuit 300 acquires the current / voltage expressed in the dq coordinate system, and acquires the first actual voltage VA1 indicating the voltage across the low-side switch element SW_A1L and the second actual voltage VA2 indicating the voltage across the low-side switch element SW_A2L. Then, the rotational speed ω of the motor is obtained. Based on the acquired current and voltage in the dq coordinate system, the first actual voltage VA1, the second actual voltage VA2, and the rotational speed ω, the control circuit 300 generates the high-side switch element SW_A1H, the high-side switch element SW_A2H, and the low-side switch element. An open failure of SW_A1L and low-side switch element SW_A2L is diagnosed.

In the present embodiment, the above-described failure diagnosis need not be performed for all three phases, and the failure diagnosis may be performed only for one phase or two phases. For example, when failure diagnosis is performed only for the A phase, only the process related to the A phase among the processes described with reference to FIGS. 5 to 17 is performed, and the processes related to the B phase and the C phase may not be performed.

(Embodiment 2)
FIG. 32 schematically shows a typical configuration of the electric power steering apparatus 3000 according to the present embodiment.

A vehicle such as an automobile generally has an electric power steering device. The electric power steering apparatus 3000 according to the present embodiment includes a steering system 520 and an auxiliary torque mechanism 540 that generates auxiliary torque. The electric power steering device 3000 generates auxiliary torque that assists the steering torque of the steering system that is generated when the driver operates the steering wheel. The burden on the driver's operation is reduced by the auxiliary torque.

The steering system 520 includes, for example, a steering handle 521, a steering shaft 522, universal shaft joints 523A and 523B, a rotation shaft 524, a rack and pinion mechanism 525, a rack shaft 526, left and right ball joints 552A and 552B, tie rods 527A and 527B, a knuckle 528A, 528B and left and right steering wheels 529A, 529B.

The auxiliary torque mechanism 540 includes, for example, a steering torque sensor 541, an automotive electronic control unit (ECU) 542, a motor 543, a speed reduction mechanism 544, and the like. The steering torque sensor 541 detects the steering torque in the steering system 520. The ECU 542 generates a drive signal based on the detection signal of the steering torque sensor 541. The motor 543 generates an auxiliary torque corresponding to the steering torque based on the dynamic signal. The motor 543 transmits the generated auxiliary torque to the steering system 520 via the speed reduction mechanism 544.

The ECU 542 includes, for example, the controller 340 and the drive circuit 350 according to the first embodiment. In an automobile, an electronic control system with an ECU as a core is constructed. In the electric power steering apparatus 3000, for example, a motor drive unit is constructed by the ECU 542, the motor 543, and the inverter 545. The motor module 2000 according to the first embodiment can be suitably used for the system.

The embodiment of the present disclosure is also suitably used for motor control systems such as X-by-wire such as shift-by-wire, steering-by-wire, and brake-by-wire, and a traction motor. For example, an EPS that implements a fault diagnosis method according to an embodiment of the present disclosure is an autonomous driving vehicle that corresponds to levels 0 to 5 (standards for automation) defined by the Japanese government and the US Department of Transportation's Road Traffic Safety Administration (NHTSA). Can be mounted.

The embodiment of the present disclosure can be widely used in various devices including various motors such as a vacuum cleaner, a dryer, a ceiling fan, a washing machine, a refrigerator, and an electric power steering device.

Claims (18)

1. A failure diagnosis method for diagnosing a failure of a power conversion device that converts power from a power source into power supplied to a motor having at least one phase winding,
   The power converter is
   At least one H-bridge, each having a first high-side switch element, a first low-side switch element, a second high-side switch element, and a second low-side switch element;
   With
   The failure diagnosis method includes:
   Determining whether there is a failed part in the high side and low side of the at least one H-bridge;
   Determining whether there is a failed set in the set of the first low-side switch element and the second high-side switch element and the set of the second low-side switch element and the first high-side switch element;
   Based on the determination result of whether there is the failed part and the determination result of whether there is the failed set, the first high-side switch element, the first low-side switch element, the second high-side switch element, and the first Determining whether there is a failed switch element among the two low-side switch elements;
   A failure diagnosis method comprising:
2. A first actual voltage indicating a voltage across the first low-side switch element; a second actual voltage indicating a voltage across the second low-side switch element; a saturation voltage; and a d-axis voltage and a q-axis voltage in a dq coordinate system. The fault diagnosis method according to claim 1, further comprising obtaining a voltage peak value determined based on the voltage peak value.
3. 3. The failure diagnosis method according to claim 2, wherein the saturation voltage is determined based on a d-axis current, a q-axis current in the dq coordinate system, and a rotation speed of the motor.
4. The saturation voltage is determined using a look-up table that associates an input of a current value determined based on the d-axis current and the q-axis current and an input of a rotation speed of the motor with the saturation voltage. Or the failure diagnosis method according to 3.
5. 3. In the step of determining whether there is a failed part, it is determined whether there is the failed part based on the first actual voltage, the second actual voltage, the saturation voltage, and the voltage peak value. 5. The failure diagnosis method according to any one of 4.
6. The obtaining step obtains a voltage command value indicating a voltage value of a target to be applied to the at least one phase winding during control of the motor and a current peak command value indicating a peak value of a current amplitude in the dq coordinate system. The fault diagnosis method according to claim 2, comprising a step.
7. The fault diagnosis method according to claim 6, wherein in the step of determining whether or not there is a faulty group, it is determined whether or not there is a faulty group based on the voltage peak value, the voltage command value, and the current peak command value. .
8. The motor has the winding of n phase (n is an integer of 3 or more),
   The fault diagnosis method according to claim 1, wherein the power conversion device includes n H bridges.
9. In the step of determining whether there is a failed part, it is determined whether there is a failed part in the high side and low side of each of the n H bridges;
   In the step of determining whether there is the failed set, the set of the first low-side switch element and the second high-side switch element of each of the n H bridges, and the second low-side switch element and the first high-side switch element The fault diagnosis method according to claim 8, wherein it is determined whether there is a faulty set in the set with the side switch element.
10. In the step of determining whether there is a failed part, if it is determined that there is a failed part before determining whether there is a failure in all of the high-side and low-side parts of the H-bridge, The failure diagnosis method according to claim 8, wherein the presence / absence determination is not performed.
11. In the step of determining whether or not there is a faulty set, if it is determined that there is a faulty set before determining the presence or absence of failure of all the sets, determination of the presence or absence of faults of the remaining sets is not performed. Item 11. The failure diagnosis method according to Item 8 or 10.
12. If it is determined that there is a failure in the step of determining whether there is a failed part,
    12. The fault diagnosis method according to claim 8, wherein the step of determining whether or not there is a faulty set is performed for 2n-1 sets of the n H bridges.
13. When it is determined that there is a failure in the step of determining whether there is a failed set,
    The failure according to any one of claims 8, 10 to 12, wherein the step of determining whether or not there is the failed part is performed for 2n-1 parts of the high side and the low side of the n H bridges. Diagnosis method.
14. The step of determining whether or not there is a failed switch element is based on a determination result of whether or not there is a failed part and a determination result of whether or not there is a failed set, the first high side switch element, the first low side The failure diagnosis method according to claim 1, further comprising a step of determining which one of a switch element, the second high-side switch element, and the second low-side switch element has failed.
15. The fault diagnosis method according to claim 14, further comprising a step of outputting a fault signal indicating the faulty switch element when the faulty switch element is specified.
16. A power conversion device that converts electric power from a power source into electric power to be supplied to a motor having at least one phase winding,
    The power converter is
    At least one H-bridge, each having a first high-side switch element, a first low-side switch element, a second high-side switch element, and a second low-side switch element;
    A control circuit for controlling the operation of the at least one H-bridge;
    With
    The control circuit includes:
    Determining if there is a failed part in the high side and low side of the at least one H-bridge;
    Determining whether there is a failed set in the set of the first low-side switch element and the second high-side switch element and the set of the second low-side switch element and the first high-side switch element;
    Based on the determination result of whether there is the failed part and the determination result of whether there is the failed set, the first high-side switch element, the first low-side switch element, the second high-side switch element, and the first 2. A power conversion device that determines whether there is a failed switch element among the low-side switch elements.
17. A motor,
    The power conversion device according to claim 16,
    A motor module comprising:
18. An electric power steering apparatus comprising the motor module according to claim 17.

Images