Switching tube fault diagnosis method based on phase current per-unit average value

Abstract

The invention discloses a switching tube fault diagnosis method based on a phase current per-unit average value, relates to the field of aviation motor control, is applied to an open winding SDSEM driving system, and comprises the following steps: determining the phase current per-unit average value of a p-phase armature winding in a nearest electric period interval of a current electric angle position; positioning a fault switch tube pair connected with a p-phase armature winding according to the detected per-unit average value of the phase current; in a conduction sector of a non-fault switch tube pair connected with the p-phase armature winding, two switch tubes in the non-fault switch tube pair are kept to be switched off, and a current path is constructed by utilizing switch tubes in a fault switch tube pair connected with the p-phase armature winding to inject a diagnosis pulse to the p-phase armature winding; and determining a switch tube with an open-circuit fault in the fault switch tube pair according to the response current of the p-phase armature winding after the diagnosis pulse is injected. And the open-circuit fault positioning of a single tube and double tubes is realized through the combination of the per-unit average value of the phase current and the diagnosis pulse injection.

Description

Technical Field

[0001] The present application relates to the field of aviation motor control, and in particular to a switch tube fault diagnosis method based on the per-unit average value of phase current. Background Art

[0002] Sinusoidal Doubly Salient Electromagnetic Machine (SDSEM) is a new type of reluctance motor with the advantages of simple rotor structure, low manufacturing cost, strong high-temperature and high-speed operation capability, and timely demagnetization in case of fault. Its torque pulsation is smaller than that of traditional electrically excited doubly salient motor. It has broad application prospects in high-requirement fields such as aviation, new energy power generation, and electric vehicles.

[0003] As the core component of the sinusoidal electrically excited doubly salient motor drive system, the main power converter has a stable and reliable operating performance, which is the key factor to ensure the stable operation of the sinusoidal electrically excited doubly salient motor. As an important carrier for implementing various control strategies, the main power converter often carries high voltage and high current in the system, and is a weak link in the system that is prone to failure. In particular, the open circuit failure of the switch tube in the main power converter seriously affects the performance of the motor drive system and even damages the sinusoidal electrically excited doubly salient motor.

[0004] The traditional method of open-circuit fault diagnosis based on the per-unit average value of phase current can only realize single-tube open-circuit fault diagnosis. The limitations of this method lead to low accuracy of the diagnostic results, which affects the operating safety of the sinusoidal electrically excited doubly salient pole motor. Summary of the invention

[0005] In view of the above problems and technical requirements, the present application proposes a switch tube fault diagnosis method based on the per-unit average value of phase current, which is applied to an open-winding SDSEM drive system. The main power converter in the open-winding SDSEM drive system includes two three-bridge-arm inverters. One end of the three-phase armature windings A, B, and C of the open-winding SDSEM is respectively connected to the midpoints of the three first bridge arms in a three-bridge-arm inverter, and the other end of the three-phase armature windings A, B, and C is respectively connected to the midpoints of the three second bridge arms in another three-bridge-arm inverter. The technical solution of the present application is as follows:

[0006] A method for diagnosing switch tube faults based on per-unit average value of phase currents comprises the following steps:

[0007] Determine the per-unit average value D of the phase current of the p-phase armature winding in the most recent electrical cycle interval at the current electrical angle position θ p , where p = a, b, c represent the armature windings of phases A, B, and C respectively;

[0008] After the phase current per unit average value D is detected p The absolute value of |D p |When the error range is within 1, it is determined that there is an open circuit fault in the switch tube in the two bridge arms connected to the p-phase armature winding; according to the average value of the phase current D p The positive or negative value of locates the faulty switch tube pair connected to the p-phase armature winding, each switch tube pair includes an upper bridge arm switch tube in one bridge arm connected to the p-phase armature winding and a lower bridge arm switch tube in another bridge arm;

[0009] In the conduction sector of the non-fault switch tube pair connected to the p-phase armature winding, the two switch tubes in the non-fault switch tube pair are kept turned off, and the switch tubes in the faulty switch tube pair connected to the p-phase armature winding are used to build a current path to inject a diagnostic pulse into the p-phase armature winding, and the switch tube with an open circuit fault in the faulty switch tube pair is determined based on the response current of the p-phase armature winding after the diagnostic pulse is injected.

[0010] A further technical solution is that injecting a diagnostic pulse into the p-phase armature winding comprises:

[0011] When the faulty switch tube pair is located within the conduction sector of the faulty switch tube pair connected to the p-phase armature winding, the current operating state is maintained until entering the conduction sector of the non-faulty switch tube pair connected to the p-phase armature winding. A current path is constructed using the switch tubes in the faulty switch tube pair of the p-phase armature winding and the anti-parallel diodes of the switch tubes in the non-faulty switch tube pair to inject a diagnostic pulse into the p-phase armature winding by self-injection of back-electromotive force.

[0012] A further technical solution is that injecting a diagnostic pulse into the p-phase armature winding comprises:

[0013] When the faulty switch tube pair is located within the conduction sector of the non-faulty switch tube pair connected to the p-phase armature winding, and when it is determined based on the fault detection position θ0 that the remaining conduction width of the non-faulty switch tube pair in the current conduction sector meets the diagnostic pulse width requirement, a current path is constructed using the switch tubes in the faulty switch tube pair of the p-phase armature winding and the anti-parallel diodes of the switch tubes of the non-faulty switch tube pair to inject a diagnostic pulse into the p-phase armature winding through self-injection of back-electromotive force.

[0014] A further technical solution is that injecting a diagnostic pulse into the p-phase armature winding comprises:

[0015] When a faulty switch tube pair is located within the conduction sector of a non-faulty switch tube pair connected to the p-phase armature winding, and it is determined based on the fault detection position θ0 that the remaining conduction width of the non-faulty switch tube pair in the current conduction sector does not meet the diagnostic pulse width requirement, the current operating state is maintained until the next conduction sector of the non-faulty switch tube pair, and a current path is constructed using the switch tubes in the faulty switch tube pair of the p-phase armature winding and the anti-parallel diodes of the switch tubes of the non-faulty switch tube pair to inject a diagnostic pulse into the p-phase armature winding by self-injection of back-electromotive force.

[0016] A further technical solution is that the midpoint of each first bridge arm in the open-winding SDSEM is bridged to the midpoint of a second bridge arm through a bidirectional switch tube, and the switch state of the upper bridge arm switch tube of the first bridge arm bridged by the same bidirectional switch tube is the same as the switch state of the upper bridge arm switch tube of the second bridge arm bridged by it, and the switch state of the lower bridge arm switch tube of the first bridge arm bridged by the same bidirectional switch tube is the same as the switch state of the lower bridge arm switch tube of the second bridge arm bridged by it;

[0017] The diagnostic pulse injection into the p-phase armature winding includes:

[0018] When a faulty switch tube pair is located within the conduction sector of the non-faulty switch tube pair connected to the p-phase armature winding, and it is determined based on the fault detection position θ0 that the remaining conduction width of the non-faulty switch tube pair in the current conduction sector does not meet the diagnostic pulse width requirement, a current path is constructed using the switch tubes in the faulty switch tube pair of the p-phase armature winding and the bidirectional switch tubes connected to the p-phase armature winding, and a diagnostic pulse is injected into the p-phase armature winding through main power injection.

[0019] A further technical solution is that injecting a diagnostic pulse into the p-phase armature winding by back-electromotive force self-injection includes:

[0020] The two switch tubes in the fault switch tube pair connected to the p-phase armature winding are controlled to conduct one tube at a time. When the response current of the p-phase armature winding in the process of single tube conduction reaches the threshold value i th When the faulty switch tube is detected, it is determined that the switched-off switch tube has an open circuit fault and the switched-on switch tube is operating normally;

[0021] When the response current of the p-phase armature winding during the single-tube conduction process does not reach the threshold value i th When , it is determined that both switch tubes in the faulty switch tube pair have open circuit faults.

[0022] A further technical solution is that injecting a diagnostic pulse into the p-phase armature winding by main power injection includes:

[0023] The two switch tubes in the fault switch tube pair connected to the p-phase armature winding are controlled to be turned on in sequence, and when one of the switch tubes in the fault switch tube pair is turned on, the bidirectional switch tube connected to the bridge arm where the other switch tube is located is controlled to be turned on, and the other two bidirectional switch tubes are kept turned off, so as to form a current path among the main power supply, the turned-on switch tube in the turned-on fault switch tube pair, the p-phase armature winding, the turned-on bidirectional switch tube, and the switch tube in the main power converter with the same switching state as the turned-off switch tube in the turned-on fault switch tube pair, and inject a diagnostic pulse. When the response current of the p-phase armature winding in the process of single-tube conduction reaches a threshold value i th When the faulty switch tube is detected, it is determined that the switched-off switch tube has an open circuit fault and the switched-on switch tube is operating normally;

[0024] When the response current of the p-phase armature winding during the single-tube conduction process does not reach the threshold value i th When , it is determined that both switch tubes in the faulty switch tube pair have open circuit faults.

[0025] A further technical solution is to take the direction from the midpoint of the first bridge arm connected to the p-phase armature winding to the midpoint of the second bridge arm as the positive direction of the current, and according to the per-unit average value D of the phase current p The positive and negative values ​​of the faulty switch pairs connected to the p-phase armature winding include:

[0026] When the phase current is per unit average value D p When the error is within the range of -1, it is determined that the switch tube pair consisting of the upper switch tube of the first bridge arm and the lower switch tube of the second bridge arm connected to the p-phase armature winding is a faulty switch tube pair;

[0027] When the phase current is per unit average value D p When the error is within the range of 1, it is determined that the switch tube pair consisting of the lower switch tube of the first bridge arm and the upper switch tube of the second bridge arm connected to the p-phase armature winding is a faulty switch tube pair.

[0028] A further technical solution is to detect whether the remaining conduction width of the non-faulty switch tube pair in the current conduction sector meets the diagnostic pulse width requirement, including:

[0029] When it is detected that the fault detection position θ0 is at least as far away from the end position of the current conduction sector of the non-fault switch pair connected to the p-phase armature winding as the diagnostic pulse injection width θ2 of the back-electromotive force self-injection method, it is determined that the remaining conduction width of the non-fault switch pair in the current conduction sector meets the diagnostic pulse width requirement; otherwise, it is determined that the remaining conduction width of the non-fault switch pair in the current conduction sector does not meet the diagnostic pulse width requirement.

[0030] Its further technical solution is that, according to the characteristics of instantaneous injection of diagnostic pulses by the main power injection method, the diagnostic pulse injection width θ4 of the main power injection method is smaller than the diagnostic pulse injection width θ2 of the back-electromotive force self-injection method; when the remaining conduction width of the non-fault switch tube pair in the current conduction sector is less than the diagnostic pulse injection width θ2 of the back-electromotive force self-injection method, a current path is constructed by using the switch tube in the faulty switch tube pair of the p-phase armature winding and the bidirectional switch tube connected to the p-phase armature winding, and the diagnostic pulse is injected into the p-phase armature winding by the main power injection method.

[0031] The beneficial technical effects of this application are:

[0032] The present application proposes a method for diagnosing switch tube faults based on the per-unit average value of phase current. By calculating the average value of phase current within an electrical cycle interval and performing per-unit processing, it can realize open circuit fault detection and determine the faulty phase and faulty switch tube pair, which can avoid misdiagnosis caused by too small load current under light load or no-load conditions, and effectively improve the robustness and anti-interference of the fault diagnosis method. The method of constructing a current path and injecting diagnostic pulses using the switch tubes in the faulty switch tube pair can accurately locate the switch tube in the faulty switch tube pair that has an open circuit fault. Compared with the traditional fault diagnosis method of the per-unit average value of phase current, it can not only accurately locate single-tube faults, but also accurately locate double-tube faults.

[0033] By using the motor's back EMF to self-inject diagnostic pulses, on the basis of realizing single-tube and dual-tube fault detection, only the motor back EMF of the open-winding SDSEM drive system itself is used to accurately locate the specific switch tube with an open circuit fault, and the speed of fault location is effectively improved. By analyzing the fault detection location to determine the diagnostic pulse injection method, and combining the back EMF self-injection method with the main power injection method, the speed of fault location is further improved, ensuring that the fault location is completed within one electrical cycle interval, and solving the problem of misdiagnosis caused by the failure of fault location feature extraction due to the long fault detection time of the traditional phase current per-unit average fault diagnosis method. The method of the present application is simple to implement and low in cost, and can fully utilize the advantages of different diagnostic pulse injection methods to significantly improve the speed of fault location. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is the topological structure diagram of the open-winding SDSEM drive system.

[0035] Figure 2 It is a flow chart of the switch tube fault diagnosis method.

[0036] Figure 3 It is the current forward loop when the switch tube T1 fails to open circuit.

[0037] Figure 4This is a schematic diagram of the switch tube pair connected to the A-phase armature winding.

[0038] Figure 5 It is a schematic diagram of the fault detection interval and the diagnostic pulse injection interval.

[0039] Figure 6 It is the current flow path after the reverse EMF is self-injected when T1 fails to open circuit.

[0040] Figure 7 It is the current flow path after the reverse electromotive force is self-injected when T8 fails to open the circuit.

[0041] Figure 8 It is the motor back EMF and response current after back EMF self-injection when T1 has an open circuit fault.

[0042] Fig. 9 It is the motor back EMF and response current after back EMF self-injection when T8 has an open circuit fault.

[0043] Fig.10 It is the motor back EMF and response current after back EMF self-injection when T1 and T8 open circuit fault.

[0044] Fig.11 This is the topological structure diagram of the open-winding SDSEM drive system with a bidirectional switch tube added.

[0045] Fig.12 It is the current flow path after the main power is injected when T1 fails to open circuit.

[0046] Fig.13 It is the current flow path after the main power is injected when T8 fails to open the circuit. DETAILED DESCRIPTION

[0047] The specific implementation of the present application is further described below in conjunction with the accompanying drawings.

[0048] Compared with the star connection topology of the traditional sinusoidal electric excitation double-pole motor drive system, the present application proposes a switch tube fault diagnosis method based on the per-unit average value of phase current, which is applied to the open-winding SDSEM drive system. The open-winding SDSEM drive system of the present application opens the neutral point of the three-phase armature winding. The specific structure is as follows Figure 1As shown in the figure, the main power converter in the open-winding SDSEM drive system includes two three-bridge-arm inverters. One end of the three-phase armature windings A, B, and C of the open-winding SDSEM is connected to the midpoints of the three first bridge arms in a three-bridge-arm inverter, and the other end of the three-phase armature windings A, B, and C is connected to the midpoints of the three second bridge arms in another three-bridge-arm inverter. Four switch tubes are connected to the two ends of each phase armature winding, and each switch tube is anti-parallel connected to a diode. Compared with the single inverter structure of the traditional sinusoidal electric excitation double-pole motor drive system, the open-winding SDSEM drive system has better fault tolerance and flexibility. When the switch tube of a certain inverter fails, the lower bridge arms of each phase of the inverter can be turned on, while the other inverter remains normal. At this time, the dual inverter system changes to a single inverter working state, which can achieve constant torque half-speed operation, thereby ensuring the safe operation of the system.

[0049] Based on the above open-winding SDSEM drive system structure, this application proposes a method for diagnosing open-circuit faults of switch tubes in an open-winding SDSEM drive system. Please refer to Figure 2 The flowchart shown in the figure has the following specific steps:

[0050] Step 1: Sample the phase current of the p-phase armature winding in the most recent electrical cycle interval of the current electrical angle position θ by means of a sliding cycle and determine the per-unit average value D of the phase current p , where p=a, b, c represent the A, B, and C phase armature windings respectively.

[0051] When the open-winding sinusoidal electro-excited doubly salient motor power converter has a switch open-circuit fault, the power converter cannot output the entire voltage vector, causing stator current distortion. Due to the freewheeling effect of the anti-parallel diode of the switch in the three-leg inverter, the switch open-circuit fault only affects the current waveform of half an electrical cycle.

[0052] Taking the open circuit fault of switch tube T1 as an example for analysis, the direction from the midpoint of the first bridge arm connected by the p-phase armature winding to the midpoint of the second bridge arm is the positive direction of the current, and the direction from a1 to a2 connected by the A-phase armature winding is the positive direction of the current. The positive current loop formed by the A-phase armature winding connecting the switch tube is as follows: Figure 3 As shown, only the forward current loop of T1 is turned on. Figure 3 As shown in (a), when T1 has an open circuit fault, only Figure 3 (a) The forward current of the current loop shown in Figure 1 has an impact. Since T1 is open, the forward current loop is Figure 3 (a) becomes Figure 3 (d) Accelerate the decay of the forward current. When T1 and T8 are turned on at the same time, the forward current loop is Figure 3 (b) becomes Figure 3(c) , which prevents the forward current from increasing. Therefore, when the switch tube T1 fails to open, the A phase current i a There will be a more obvious negative bias, or even only negative current. Similarly, when the switch tube T8 has an open circuit fault, the A phase current i a There will also be a more obvious negative bias, or even only negative current. Similarly, when the switch tube T7 or T9 has an open circuit failure, the A phase current i a There will be a more obvious forward bias, or even only forward current.

[0053] Since the two three-bridge-arm inverters in the open-winding sinusoidal electromagnetically excited double-pole motor drive system have a certain degree of symmetry, the two switches (such as T1 and T8, T2 and T7) in the same phase that are simultaneously turned on in the positive or negative direction of the current have the same open-circuit fault characteristics. The upper bridge arm switch in one bridge arm connected to the p-phase armature winding and the lower bridge arm switch in the other bridge arm are defined as switch pairs. The switch pair connected to the A-phase armature winding is as follows: Figure 4 As shown, T1 and T8 form a switch pair, T2 and T7 form a switch pair; T3 and T7 connected to the armature winding of phase B 10 Form a switch pair, T4 and T9 form a switch pair; T5 and T 12 Form a switch pair, T6 and T 11 Form a switch tube pair.

[0054] According to the above analysis, an open circuit fault in the switch tube will cause part of the phase current waveform to be missing, causing the phase current cycle average value to be offset. Therefore, the fault detection can be achieved and the faulty phase and faulty switch tube pair can be determined by the average value of the phase current within an electrical cycle. However, since the magnitude of the phase current average value is easily affected by the load current, when the motor is lightly loaded or unloaded, the phase current is small or even close to 0, which may cause misdiagnosis. In order to improve the robustness and anti-interference ability of the fault diagnosis method, the phase current average value is normalized, and the ratio of the phase current average value within an electrical cycle interval to the average value of the phase current absolute value is defined as the phase current normalized average value. T is the length of an electrical cycle interval.

[0055] Step 2: After the phase current average value D is detected p The absolute value of |D p |When the error is within the range of 1, it is determined that there is an open circuit fault in the switch tube in the two bridge arms connected to the p-phase armature winding. p The positive or negative value locates the faulty switch tube pair connected to the p-phase armature winding.

[0056] When no fault occurs, D pThe value is 0; when an open circuit fault occurs in a switch tube, the D p The value is -1 or 1. Due to the vector control effect, the three-phase current will have a DC bias in the same direction. p The value is 0<|D p |<1. Available through D p The value of is used to detect the occurrence of open circuit fault and determine the fault phase. Considering the influence of sampling accuracy and external interference, the fault detection expression is:

[0057]

[0058] Among them, ε is an infinitesimal positive number, D a is the per-unit average value of the phase current of the A-phase armature winding, D b is the per-unit average value of the phase current of the B-phase armature winding, D c It is the per-unit average value of the phase current of the C-phase armature winding.

[0059] In one embodiment, according to the phase current per unit average value D p The positive and negative values ​​of the positioning p-phase armature winding connected to the fault switch tube pair include: when the phase current standard average value D p When the error range is -1, it is determined that the switch pair consisting of the upper switch of the first bridge arm and the lower switch of the second bridge arm connected to the p-phase armature winding is a faulty switch pair; when the per-unit average value of the phase current D p When the error is within the range of 1, it is determined that the switch tube pair consisting of the lower switch tube of the first bridge arm and the upper switch tube of the second bridge arm connected to the p-phase armature winding is a faulty switch tube pair.

[0060] Taking phase A as an example, when the switch tubes of switch tube pair T1 and T8 have an open circuit fault, the current waveform loses the positive half cycle, and D a The value is around -1; when the switch tubes T2 and T7 have an open circuit failure, the current waveform lacks the negative half cycle, D a The value is close to 1. The same is true for B and C. Therefore, according to the D of the fault phase p The positive and negative values ​​are used to determine the faulty switch pair. The switch pair with open circuit fault is p The relationship is as follows:

[0061]

[0062] When an open circuit fault occurs in the switch tubes of the same switch tube pair, the same fault current characteristics exist, and the faulty switch tube needs to be further accurately located.

[0063] Step 3, in the conduction sector of the non-fault switch tube pair connected to the p-phase armature winding, keep the two switch tubes in the non-fault switch tube pair turned off, and use the switch tubes in the faulty switch tube pair connected to the p-phase armature winding to build a current path to inject a diagnostic pulse into the p-phase armature winding, and determine the switch tube with an open circuit fault in the faulty switch tube pair based on the response current of the p-phase armature winding after the diagnostic pulse injection.

[0064] When an open-circuit fault occurs in one of the switch tubes in the switch tube pair connected to the p-phase armature winding in the forward current loop, the power supply can no longer provide positive energy to the faulty phase, so the phase current loses the positive half cycle. However, thanks to the strong fault-tolerant operation capability of the open-winding SDSEM drive system, the other two non-faulty phases can still maintain the motor operation, and the faulty phase is converted from the motoring mode to the generating mode.

[0065] In one electrical angle cycle, the conduction sectors of the faulty switch pair and the non-faulty switch pair of the faulty phase each occupy half an electrical angle cycle, and the motor back-EMF generated by the conduction of the faulty switch pair is opposite to the motor back-EMF generated by the conduction of the non-faulty switch pair, and due to the freewheeling effect of the anti-parallel diode of the switch tube, the motor back-EMF generated by the non-faulty switch tube in the conduction sector can be used to realize the self-injection of diagnostic pulses through the loop formed by the anti-parallel diode of the non-faulty switch tube and the faulty switch tube. For the case where the faulty switch tube pair is in the positive current loop, it is only necessary to provide a positive loop that satisfies the operation of any one of the switch tubes alone, and the motor back-EMF generated by the non-faulty switch tube pair in the conduction sector can be used to inject diagnostic pulses, and then observe whether a positive response current is generated, thereby locating the faulty switch tube. Conversely, for the case where the faulty switch tube pair is in the negative current loop, it is only necessary to provide a negative loop that satisfies the operation of any one of the switch tubes alone, and the motor back-EMF generated by the non-faulty switch tube pair in the conduction sector can be used to inject diagnostic pulses, and then observe whether a negative response current is generated, thereby locating the faulty switch tube.

[0066] Since the open circuit fault occurs in the conduction sector of the faulty switch pair, and the calculation of the per-unit average value of the phase current requires the sampling of the phase current within an entire electrical cycle, the fault cannot be detected immediately when the open circuit fault occurs. In addition, since the sliding cycle sampling is used to calculate the average value, the fault detection position of the open circuit fault may appear in the conduction sector of the faulty switch pair in this cycle, the conduction sector of the non-faulty switch pair, or the conduction sector of the faulty switch pair in the next cycle according to the value characteristics of the per-unit average value of the phase current. Among them, the conduction sector of the faulty switch pair refers to the conduction electrical angle interval of the faulty switch pair within an electrical cycle interval when the motor is operating normally, and the conduction sector of the non-faulty switch pair refers to the conduction electrical angle interval of the non-faulty switch pair within an electrical cycle interval when the motor is operating normally. Take the electrical cycle interval of length 2π, and the open circuit fault occurs in the interval (-π, 0) as an example. Since the fault can only be detected after the fault occurs, and the per-unit average value of the phase current needs to calculate the phase current within an electrical cycle interval, the fault detection position is within the interval (-π, 2π), such as Figure 5 The fault detection interval is divided into 4 fault detection sub-intervals. The first sub-interval is the conduction sector of the faulty switch tube pair connected to the p-phase armature winding. The second and third sub-intervals are the conduction sectors of the non-faulty switch tube pair connected to the p-phase armature winding. The fourth sub-interval is the next conduction sector of the non-faulty switch tube pair connected to the p-phase armature winding.

[0067] θ2 is the diagnostic pulse injection width of the back-electromotive force self-injection method, and the second and third sub-intervals can be divided according to θ2. When the fault detection position is within the second sub-interval, the remaining conduction width of the non-fault switch pair in the current conduction sector meets the diagnostic pulse injection width requirement of the back-electromotive force self-injection method, and when the fault detection position is within the third sub-interval, the remaining conduction width of the non-fault switch pair in the current conduction sector does not meet the diagnostic pulse injection width requirement of the back-electromotive force self-injection method. In one embodiment, detecting whether the remaining conduction width of the non-fault switch pair in the current conduction sector meets the diagnostic pulse width requirement includes: when it is detected that the distance between the fault detection position θ0 and the end position of the current conduction sector of the non-fault switch pair connected to the p-phase armature winding is not less than the diagnostic pulse injection width θ2 of the back-electromotive force self-injection method, it is determined that the remaining conduction width of the non-fault switch pair in the current conduction sector meets the diagnostic pulse width requirement, otherwise it is determined that the remaining conduction width of the non-fault switch pair in the current conduction sector does not meet the diagnostic pulse width requirement.

[0068] For the case where a faulty switch tube is located in each fault detection sub-interval, a suitable diagnostic pulse injection interval needs to be determined. In one embodiment, when a faulty switch tube pair is located in the conduction sector (first sub-interval) of the faulty switch tube pair connected to the p-phase armature winding, the current operation state is maintained until the conduction sector of the non-faulty switch tube pair connected to the p-phase armature winding is entered, and a current path is constructed by using the switch tubes in the faulty switch tube pair of the p-phase armature winding and the anti-parallel diodes of the switch tubes in the non-faulty switch tube pair to inject diagnostic pulses into the p-phase armature winding by back-electromotive force self-injection.

[0069] When the faulty switch tube pair is located within the conduction sector (the second and third sub-intervals) of the non-faulty switch tube pair connected to the p-phase armature winding, and when it is determined based on the fault detection position θ0 that the remaining conduction width of the non-faulty switch tube pair in the current conduction sector meets the diagnostic pulse width requirement, the fault detection position θ0 is in the second sub-interval at this time, and a current path is constructed using the switch tubes in the faulty switch tube pair of the p-phase armature winding and the anti-parallel diodes of the switch tubes of the non-faulty switch tube pair to inject a diagnostic pulse into the p-phase armature winding through self-injection of back-electromotive force.

[0070] When the faulty switch tube pair is located within the conduction sector (the second and third sub-intervals) of the non-faulty switch tube pair connected to the p-phase armature winding, and it is determined based on the fault detection position θ0 that the remaining conduction width of the non-faulty switch tube pair in the current conduction sector does not meet the diagnostic pulse width requirement, the fault detection position θ0 is in the third sub-interval at this time, and the current operating state is maintained until the next conduction sector of the non-faulty switch tube pair. A current path is constructed using the switch tubes in the faulty switch tube pair of the p-phase armature winding and the anti-parallel diodes of the switch tubes of the non-faulty switch tube pair to inject a diagnostic pulse into the p-phase armature winding by self-injection of back-electromotive force.

[0071] In one embodiment, injecting a diagnostic pulse into the p-phase armature winding by back-EMF self-injection includes:

[0072] The two switch tubes in the fault switch tube pair connected to the p-phase armature winding are controlled to conduct one tube at a time. When the response current of the p-phase armature winding in the process of single tube conduction reaches the threshold value i th When the faulty switch tube is detected, it is determined that the switched-off switch tube has an open circuit fault and the switched-on switch tube is operating normally;

[0073] When the response current of the p-phase armature winding during the single-tube conduction process does not reach the threshold value i th When , it is determined that both switch tubes in the faulty switch tube pair have open circuit faults.

[0074] Take the open circuit fault of the switch pair T1 and T8 connected to the A-phase armature winding as an example for analysis. When an open circuit fault occurs in the switch pair T1 and T8, the power supply cannot provide positive energy to the A-phase armature winding, and the A-phase current lacks the positive half cycle. However, thanks to the strong fault-tolerant operation capability of the open-winding SDSEM drive system, the B and C phases can still maintain the motor operation, and the A phase is converted from the motor mode to the power generation mode. Since the switch pair T1 and T8 are in the forward current loop, it is only necessary to provide a forward loop that satisfies T1 or T8 to operate alone to inject diagnostic pulses using the motor back electromotive force.

[0075] When T1 fails to open, the forward current flows in the following path: Figure 6 As shown, at this time, only T1 needs to be turned on to build the current path of the anti-parallel diode D7 of T1 and T7, so as to realize the self-injection of the back-EMF diagnostic pulse and generate a forward response current. Similarly, when T8 has an open circuit fault, the forward current flow path is as follows Figure 7 As shown, at this time, only T8 needs to be turned on to build the current path of the anti-parallel diode D2 of T8 and T2, so as to realize the self-injection of the back-EMF diagnostic pulse and generate a forward response current. When T1 and T8 have an open circuit fault, they cannot provide a forward loop that satisfies T1 or T8 to operate alone, and there is no response current after the diagnostic pulse is self-injected. Therefore, the faulty switch tube can be accurately located according to the characteristics of the response current generated after the diagnostic pulse is self-injected.

[0076] In order to prevent the negative current generated by the non-faulty switch pair conducting the main power supply from affecting the flow of the positive current generated by the back electromotive force, the non-faulty switch pair is immediately turned off when the faulty switch pair is detected. The two switch tubes of the faulty switch tube are turned on one by one in the conduction sector of the non-faulty switch tube to inject the diagnostic pulse.

[0077] Taking the faulty switch pair T1 and T8 as an example, assuming that the fault detection position θ0 = 0, the specific process of fault location of the faulty switch pair T1 and T8 is as follows:

[0078] (1) When the motor electrical angle θ is in the interval (θ0, θ0+θ1), only T8 is turned on. θ1 is the required width of the back-EMF diagnostic pulse self-injection when a single switch is turned on. If the response current is detected to reach the threshold i th When the reverse EMF is injected, T8 is immediately turned off and it is determined that the upper bridge arm switch T1 in the faulty switch pair T1 and T8 has an open circuit fault. The reverse EMF and the response current in the reverse EMF self-injection process are as follows: Figure 8 shown.

[0079] (2) When the motor electrical angle θ is in the interval (θ0+θ1,θ0+θ2), only T1 is turned on. If the response current is detected to reach the threshold value i thWhen the reverse EMF is injected, T1 is immediately turned off and it is determined that the lower bridge arm switch T8 in the faulty switch pair T1 and T8 has an open circuit fault. Fig. 9 shown.

[0080] (3) When the motor electrical angle θ = θ0 + θ2, T1 and T8 are all turned off. If the response current is detected to be less than the threshold value i in the interval (θ0, θ0 + θ2), th , it is determined that the faulty switch pair T1 and T8 has an open circuit fault. The back EMF and response current of the back EMF self-injection process are as follows: Fig.10 shown.

[0081] It should be noted that it is possible to first turn on the upper arm switch tube or the lower arm switch tube of the faulty switch tube, and it is only necessary to ensure that the two switch tubes are turned on separately in sequence.

[0082] From the above analysis, it can be seen that when the fault detection position θ0 is in the third sub-interval, the remaining conduction width of the non-fault switch pair in the current conduction sector does not meet the diagnostic pulse width requirement, and it needs to be delayed until the next conduction sector of the non-fault switch pair to achieve self-injection of the back EMF. In order to further improve the fault location speed and avoid being limited by the back EMF, when the fault detection position θ0 is in the third sub-interval, a current path can be constructed through a bidirectional switch tube and a single switch tube of the faulty switch tube pair to achieve the injection of diagnostic pulses using the main power supply. The open-winding SDSEM drive system topology with an increased switch tube is shown in Figure 1. Fig.11 shown.

[0083] Since the switch state of any bridge arm in the three-arm inverter of the open-winding SDSEM drive system can always find a bridge arm with the same switch state in another three-arm inverter. Among them, the switch state of T1 of the A-phase bridge arm in different sectors is completely consistent with T9, and the switch state of T3 of the B-phase bridge arm is completely consistent with T9. 11 The switch states in different sectors are the same. The switch states of T5 and T7 in the C-phase bridge arm are the same in different sectors. Since the bridge arms have the same switch combination, the connection between one bridge arm and the inverter is opened and connected to the bridge arm with the same switch state. And because the switch states of the upper bridge arm switch tube and the lower bridge arm switch tube of the same bridge arm are opposite, therefore, T2 and T 10 The switching states of T4 and T4 are the same in different sectors. 12 The switching states in different sectors are the same, and the switching states of T6 and T8 in different sectors are the same.

[0084] Specifically, the bridge arm midpoint of each first bridge arm in the open winding SDSEM is bridged to the bridge arm midpoint of a second bridge arm through a bidirectional switch tube, and the switch state of the upper bridge arm switch tube of the first bridge arm bridged by the same bidirectional switch tube is the same as the switch state of the upper bridge arm switch tube of the second bridge arm bridged by it. The switch state of the lower bridge arm switch tube of the first bridge arm bridged by the same bidirectional switch tube is the same as the switch state of the lower bridge arm switch tube of the second bridge arm bridged by it.

[0085] When the fault detection position θ0 is in the third sub-interval, a current path can be constructed by a bidirectional switch tube and a single switch tube of the fault switch tube pair to realize the main power supply injection of a diagnostic pulse. In one embodiment, injecting a diagnostic pulse into the p-phase armature winding includes: when the fault switch tube pair is located in the conduction sector of the non-fault switch tube pair connected to the p-phase armature winding, and when it is determined according to the fault detection position θ0 that the remaining conduction width of the non-fault switch tube pair in the current conduction sector does not meet the diagnostic pulse width requirement, a current path is constructed using the switch tube in the fault switch tube pair of the p-phase armature winding and the bidirectional switch tube connected to the p-phase armature winding, and a diagnostic pulse is injected into the p-phase armature winding by the main power supply injection method.

[0086] In one embodiment, injecting a diagnostic pulse into the p-phase armature winding by main power injection includes:

[0087] The two switch tubes in the fault switch tube pair connected to the p-phase armature winding are controlled to be turned on in sequence, and when one of the switch tubes in the fault switch tube pair is turned on, the bidirectional switch tube connected to the bridge arm where the other switch tube is located is controlled to be turned on, and the other two bidirectional switch tubes are kept turned off, so as to form a current path among the main power supply, the turned-on switch tube in the turned-on fault switch tube pair, the p-phase armature winding, the turned-on bidirectional switch tube, and the switch tube in the main power converter with the same switching state as the turned-off switch tube in the turned-on fault switch tube pair, and inject a diagnostic pulse. When the response current of the p-phase armature winding in the process of single-tube conduction reaches a threshold value i th When the faulty switch tube is detected, it is determined that the switched-off switch tube has an open circuit fault and the switched-on switch tube is operating normally;

[0088] When the response current of the p-phase armature winding during the single-tube conduction process does not reach the threshold value i th When , it is determined that both switch tubes in the faulty switch tube pair have open circuit faults.

[0089] Taking the faulty switch pair T1 and T8 as an example, when the fault detection position θ0 is in the third sub-interval, the specific process of fault location of the faulty switch pair T1 and T8 is as follows:

[0090] (1) When the motor electrical angle θ is in the range of (θ0, θ0+θ3), only T8 and the bidirectional switch tube VD are turned on. 17, forming the main power supply, the conducting switch tube T8, the p-phase armature winding, the conducting bidirectional switch tube VD 17 , the forward current path between the switch tube T9 in the main power converter having the same switching state as the switch tube T1 that is turned off in the conduction fault switch tube pair is as follows: Fig.12 θ3 is the width required for a single switch tube to conduct the main power supply to inject a diagnostic pulse. If the response current is detected to reach the threshold i th When the fault occurs, T8 is immediately turned off and it is determined that an open circuit fault occurs in the upper arm switch tube T1 of the faulty switch tube pair T1 and T8.

[0091] (2) When the motor electrical angle θ is in the range of (θ0+θ3,θ0+θ4), only T1 and the bidirectional switch VD are turned on. 19 , forming the main power supply, the conducting switch tube T1, the p-phase armature winding, the conducting bidirectional switch tube VD 19 , the forward current path between the switch tube T6 in the main power converter having the same switching state as the switch tube T8 that is turned off in the on-fault switch tube pair is as follows: Fig.13 As shown. If the response current is detected to reach the threshold value i th When the fault occurs, T1 is immediately turned off and it is determined that an open circuit fault occurs in the lower arm switch tube T8 of the faulty switch tube pair T1 and T8.

[0092] (3) When the motor electrical angle θ = θ0 + θ4, T1 and T8 are all turned off. If the response current is detected to be less than the threshold value i in the interval (θ0, θ0 + θ4), th , it is determined that the faulty switch tube pair T1 and T8 has an open circuit fault.

[0093] Since the voltage of the main power supply is very large, the diagnostic pulse injection by the main power supply injection method has the characteristic of instantaneous injection, so the diagnostic pulse injection width θ4 of the main power supply injection method is much smaller than the diagnostic pulse injection width θ2 of the back-EMF self-injection method. Therefore, when the remaining conduction width of the non-fault switch tube pair in the current conduction sector is less than the diagnostic pulse injection width θ2 of the back-EMF self-injection method, the switch tube in the faulty switch tube pair of the p-phase armature winding and the bidirectional switch tube connected to the p-phase armature winding are used to construct a current path and inject diagnostic pulses into the p-phase armature winding by the main power supply injection method. This allows the fault location to be completed within one electrical cycle interval, solving the problem of slow diagnostic speed due to the fact that the conventional phase current per-unit average fault diagnosis requires at least two electrical cycle intervals.

[0094] The above is only a preferred embodiment of the present application, and the present application is not limited to the above embodiments. It is understood that other improvements and changes directly derived or associated by those skilled in the art without departing from the spirit and concept of the present application should be considered to be included in the protection scope of the present application.

Claims

1. A method for diagnosing switch tube faults based on the per-unit average value of phase current, characterized in that: Applied to an open-winding SDSEM drive system, the main power converter in the open-winding SDSEM drive system includes two three-bridge-arm inverters, one end of the three-phase armature windings A, B, and C of the open-winding SDSEM is respectively connected to the midpoint of the three first bridge arms in one three-bridge-arm inverter, and the other end of the three-phase armature windings A, B, and C is respectively connected to the midpoint of the three second bridge arms in another three-bridge-arm inverter. The switch tube fault diagnosis method includes: Determine the per-unit average value D of the phase current of the p-phase armature winding in the most recent electrical cycle interval at the current electrical angle position θ p , where p = a, b, c represent the armature windings of phases A, B, and C respectively; After the phase current per unit average value D is detected p The absolute value of |D p |When the error range is within 1, it is determined that there is an open circuit fault in the switch tube in the two bridge arms connected to the p-phase armature winding; according to the average value of the phase current D p The positive or negative value of locates the faulty switch tube pair connected to the p-phase armature winding, each switch tube pair includes an upper bridge arm switch tube in one bridge arm connected to the p-phase armature winding and a lower bridge arm switch tube in another bridge arm; In the conduction sector of the non-fault switch tube pair connected to the p-phase armature winding, the two switch tubes in the non-fault switch tube pair are kept turned off, and the switch tubes in the faulty switch tube pair connected to the p-phase armature winding are used to build a current path to inject a diagnostic pulse into the p-phase armature winding, and the switch tube with an open circuit fault in the faulty switch tube pair is determined based on the response current of the p-phase armature winding after the diagnostic pulse is injected.

2. The switch tube fault diagnosis method according to claim 1, characterized in that: The diagnostic pulse injection into the p-phase armature winding includes: When the faulty switch tube pair is located within the conduction sector of the faulty switch tube pair connected to the p-phase armature winding, the current operating state is maintained until entering the conduction sector of the non-faulty switch tube pair connected to the p-phase armature winding. A current path is constructed using the switch tubes in the faulty switch tube pair of the p-phase armature winding and the anti-parallel diodes of the switch tubes in the non-faulty switch tube pair to inject a diagnostic pulse into the p-phase armature winding by self-injection of back-electromotive force.

3. The switch tube fault diagnosis method according to claim 1, characterized in that: The diagnostic pulse injection into the p-phase armature winding includes: When the faulty switch tube pair is located within the conduction sector of the non-faulty switch tube pair connected to the p-phase armature winding, and when it is determined based on the fault detection position θ0 that the remaining conduction width of the non-faulty switch tube pair in the current conduction sector meets the diagnostic pulse width requirement, a current path is constructed using the switch tubes in the faulty switch tube pair of the p-phase armature winding and the anti-parallel diodes of the switch tubes of the non-faulty switch tube pair to inject a diagnostic pulse into the p-phase armature winding through self-injection of back-electromotive force.

4. The switch tube fault diagnosis method according to claim 1, characterized in that: The diagnostic pulse injection into the p-phase armature winding includes: When a faulty switch tube pair is located within the conduction sector of a non-faulty switch tube pair connected to the p-phase armature winding, and it is determined based on the fault detection position θ0 that the remaining conduction width of the non-faulty switch tube pair in the current conduction sector does not meet the diagnostic pulse width requirement, the current operating state is maintained until the next conduction sector of the non-faulty switch tube pair, and a current path is constructed using the switch tubes in the faulty switch tube pair of the p-phase armature winding and the anti-parallel diodes of the switch tubes of the non-faulty switch tube pair to inject a diagnostic pulse into the p-phase armature winding by self-injection of back-electromotive force.

5. The switch tube fault diagnosis method according to claim 1, characterized in that: The bridge arm midpoint of each first bridge arm in the open winding SDSEM is bridged to the bridge arm midpoint of a second bridge arm through a bidirectional switch tube, and the switch state of the upper bridge arm switch tube of the first bridge arm bridged by the same bidirectional switch tube is the same as the switch state of the upper bridge arm switch tube of the second bridge arm bridged by it, and the switch state of the lower bridge arm switch tube of the first bridge arm bridged by the same bidirectional switch tube is the same as the switch state of the lower bridge arm switch tube of the second bridge arm bridged by it; The diagnostic pulse injection into the p-phase armature winding includes: When a faulty switch tube pair is located within the conduction sector of the non-faulty switch tube pair connected to the p-phase armature winding, and it is determined based on the fault detection position θ0 that the remaining conduction width of the non-faulty switch tube pair in the current conduction sector does not meet the diagnostic pulse width requirement, a current path is constructed using the switch tubes in the faulty switch tube pair of the p-phase armature winding and the bidirectional switch tubes connected to the p-phase armature winding, and a diagnostic pulse is injected into the p-phase armature winding through main power injection.

6. The switch tube fault diagnosis method according to any one of claims 2 to 4, characterized in that: The diagnostic pulses injected into the p-phase armature winding by back EMF self-injection include: The two switch tubes in the fault switch tube pair connected to the p-phase armature winding are controlled to conduct one tube at a time. When the response current of the p-phase armature winding in the process of single tube conduction reaches the threshold value i th When the faulty switch tube is detected, it is determined that the switched-off switch tube has an open circuit fault and the switched-on switch tube is operating normally; When the response current of the p-phase armature winding during the single-tube conduction process does not reach the threshold value i th When , it is determined that both switch tubes in the faulty switch tube pair have open circuit faults.

7. The switch tube fault diagnosis method according to claim 5, characterized in that: The diagnostic pulses injected into the p-phase armature winding by main power injection include: The two switch tubes in the fault switch tube pair connected to the p-phase armature winding are controlled to be turned on in sequence, and when one of the switch tubes in the fault switch tube pair is turned on, the bidirectional switch tube connected to the bridge arm where the other switch tube is located is controlled to be turned on, and the other two bidirectional switch tubes are kept turned off, so as to form a current path among the main power supply, the turned-on switch tube in the turned-on fault switch tube pair, the p-phase armature winding, the turned-on bidirectional switch tube, and the switch tube in the main power converter with the same switching state as the turned-off switch tube in the turned-on fault switch tube pair, and inject a diagnostic pulse. When the response current of the p-phase armature winding in the process of single-tube conduction reaches a threshold value i th When the faulty switch tube is detected, it is determined that the switched-off switch tube has an open circuit fault and the switched-on switch tube is operating normally; When the response current of the p-phase armature winding during the single-tube conduction process does not reach the threshold value i th When , it is determined that both switch tubes in the faulty switch tube pair have open circuit faults.

8. The switch tube fault diagnosis method according to claim 1, characterized in that: The direction from the midpoint of the first bridge arm connected to the p-phase armature winding to the midpoint of the second bridge arm is the positive direction of the current. According to the per-unit average value of the phase current D p The positive and negative values ​​of the faulty switch tube pairs connected to the p-phase armature winding include: When the phase current per unit average value D p When the error is within the range of -1, it is determined that the switch tube pair consisting of the upper switch tube of the first bridge arm and the lower switch tube of the second bridge arm connected to the p-phase armature winding is a faulty switch tube pair; When the phase current per unit average value D p When the error is within the range of 1, it is determined that the switch tube pair consisting of the lower switch tube of the first bridge arm and the upper switch tube of the second bridge arm connected to the p-phase armature winding is a faulty switch tube pair.

9. The switch tube fault diagnosis method according to any one of claims 3 to 5, characterized in that: Detecting whether the remaining conduction width of the non-faulty switch tube pair in the current conduction sector meets the diagnostic pulse width requirement includes: When it is detected that the fault detection position θ0 is at least as far away from the end position of the current conduction sector of the non-fault switch pair connected to the p-phase armature winding as the diagnostic pulse injection width θ2 of the back-electromotive force self-injection method, it is determined that the remaining conduction width of the non-fault switch pair in the current conduction sector meets the diagnostic pulse width requirement; otherwise, it is determined that the remaining conduction width of the non-fault switch pair in the current conduction sector does not meet the diagnostic pulse width requirement.

10. The switch tube fault diagnosis method according to claim 9, characterized in that: According to the characteristics of the instantaneous injection of diagnostic pulses by the main power injection method, the diagnostic pulse injection width θ4 of the main power injection method is smaller than the diagnostic pulse injection width θ2 of the back-EMF self-injection method. When the remaining conduction width of the non-faulty switch tube pair in the current conduction sector is less than the diagnostic pulse injection width θ2 of the back-EMF self-injection method, a current path is constructed using the switch tube in the faulty switch tube pair of the p-phase armature winding and the bidirectional switch tube connected to the p-phase armature winding, and the diagnostic pulse is injected into the p-phase armature winding by the main power injection method.

Images