Switched reluctance motor drive system multi-fault diagnosis method based on current behavior

Abstract

The application provides a kind of switch reluctance motor drive system multi-fault diagnosis method based on current behavior, belong to motor drive system fault diagnosis field.The method reselects the installation position of the three current sensors inherent to motor without increasing the number of sensors, by changing the measurement path of each sensor and the number of measurement current branches, accurately reconstructing the three-phase current value, after the occurrence of fault, the monitoring current value of the corresponding device and the theoretical current value appear larger deviation, by analyzing this current deviation behavior, the rapid diagnosis and positioning of the multiple component faults of switch reluctance motor drive system can be realized, including current sensor fault, winding open circuit fault, power diode open circuit fault and switch tube open circuit short circuit mixed fault, greatly facilitate the monitoring of the health status of switch reluctance motor drive system.

Description

Technical Field

[0001] This invention belongs to the field of fault diagnosis technology for motor drive systems, and specifically relates to a multi-fault diagnosis method for switched reluctance motor drive systems based on current behavior. Background Technology

[0002] Compared to traditional motors, switched reluctance motors (SRMs) offer advantages such as simple structure, high efficiency, flexible control, and low cost, leading to their widespread application in specific fields such as electric vehicles, home appliances, aerospace, and industrial automation equipment. However, the unique magnetic circuit saturation effect, strong nonlinear characteristics, and pulsed power supply of SRMs result in complex fault coupling characteristics in their drive systems when faced with power converter faults (such as IGBT open / short circuits or freewheeling diode failures) and winding faults (inter-turn short circuits or phase-to-phase short circuits). This poses a significant challenge to online fault diagnosis technology.

[0003] In existing technologies, fault diagnosis methods based on current signal analysis have become a research hotspot because they directly reflect the operating status and health of motor drive systems. Traditional solutions typically employ three current sensors configured in the motor winding circuit. However, this architecture has significant technical limitations: First, for a typical three-phase switched reluctance motor paired with a three-phase asymmetrical half-bridge power converter, which includes 12 independent current branches consisting of 6 IGBT power switches and 6 freewheeling diodes, the existing sensor arrangement can only acquire the total current information of the three-phase windings, failing to directly observe the current behavior characteristics of each power device's branch. This results in some device faults going undetected or fault detection taking a long time, affecting the operation of the entire electric drive system. Second, when a single branch component fails, the fault characteristic signal is masked by the normal branch current, leading to insufficient sensitivity of existing methods for early, weak faults and difficulty in accurately locating the faulty component. Therefore, existing solutions often require a separate current sensor for each power branch, significantly increasing equipment cost and weight. Summary of the Invention

[0004] The purpose of this invention is to solve the aforementioned problems existing in the prior art, and to provide a multi-fault diagnosis method for switched reluctance motor drive systems based on current behavior. By changing the installation position of the sensors without increasing the number of current sensors, each power device in the drive circuit can be monitored in real time. Furthermore, a multi-fault diagnosis method combined with the installation method is proposed. By measuring and calculating current behavior in real time, the health status of the drive system is determined, thereby achieving rapid fault diagnosis and location.

[0005] To achieve the above objectives, the technical solution provided by this invention is:

[0006] A multi-fault diagnosis method for switched reluctance motor drive systems based on current behavior is provided, including the following steps:

[0007] Step 1: Install a first current sensor, a second current sensor, and a third current sensor in the switched reluctance motor drive system; wherein, each phase drive circuit of the switched reluctance motor consists of winding branches LA, LB, LC, and power device branches on both sides of the half-bridge. The two switching transistors and two power diodes on both sides of each phase form an H-bridge. Each phase includes a left-side switching transistor branch xLH, a left-side power diode branch xLL, a right-side switching transistor branch xRL, and a right-side power diode branch xRH, where x represents phase A, B, or C;

[0008] The first current sensor is connected across the common node of the C-phase right-side switch branch CRL, the A-phase winding branch LA, and the B-phase left-side switch branch BLH, and is used to monitor their combined current; the second current sensor is connected across the common node of the A-phase right-side switch branch ARL, the B-phase winding branch LB, and the C-phase left-side switch branch CLH, and is used to monitor their combined current; the third current sensor is connected across the common node of the B-phase right-side switch branch BRL, the C-phase winding branch LC, and the A-phase left-side switch branch ALH, and is used to monitor their combined current.

[0009] Step 2: Collect the current values ​​measured by the first current sensor, the second current sensor, and the third current sensor, and calculate the winding phase current value based on the collected current sensor current values;

[0010] Step 3: Set the current sensor fault diagnosis variable according to the collected current sensor current value, and determine whether the first current sensor, the second current sensor and the third current sensor are abnormal according to the current sensor fault diagnosis variable. If it is determined that there may be an abnormality, proceed to step 4; if it is determined that there is no abnormality, proceed to step 5.

[0011] Step 4: Continuously monitor the current sensor fault diagnosis variables at two different times. If both values ​​are close to zero, it is determined that a current sensor fault has occurred, and proceed to step 8. If no values ​​close to zero are continuously observed, it is determined that no current sensor fault has occurred, and proceed to step 5.

[0012] Step 5: Set phase current fault diagnosis variables according to the solved winding phase current values, and determine whether the first current sensor, the second current sensor and the third current sensor are abnormal based on the phase current fault diagnosis variables. If an abnormality is found, proceed to step 6; if no abnormality is found, proceed to step 7.

[0013] Step 6: Determine whether the phase current fault diagnosis variable is normal at the next moment. If it is normal, it is determined that a diode open circuit fault has occurred, and proceed to step 8. If it is still abnormal, it is determined that a winding open circuit fault has occurred, and proceed to step 8.

[0014] Step 7: Set the fault diagnosis variable for the switch tube branch current based on the collected current sensor current value and the solved winding phase current value, and determine whether a switch tube branch current fault has occurred based on the fault diagnosis variable for the switch tube branch current. If a switch tube open circuit or short circuit fault occurs, proceed to step 8. If no fault occurs, return to step 2 to continue fault diagnosis.

[0015] Step 8: Output the fault diagnosis results.

[0016] Furthermore, in step 2, the winding phase current is solved according to the following sub-steps:

[0017] Step 2.1: Establish the current equations measured by the first current sensor, the second current sensor, and the third current sensor:

[0018]

[0019] In the formula, i1, i2, and i3 are the phase currents of the three-phase windings, respectively, and the measured values ​​of the first, second, and third current sensors are the measured values ​​of the first, second, and third current sensors, respectively. These represent the on / off status of switching transistors S1, S2, S3, S4, S5, and S6, respectively.

[0020] Step 2.2: Solve for the winding phase currents based on the established current equations:

[0021] .

[0022] Furthermore, , , , , , The value is determined as follows:

[0023]

[0024] In the formula, m= , , , , , .

[0025] Furthermore, in step 3, the current sensor fault diagnosis variables are set as follows:

[0026]

[0027] In the formula, and These are the current values ​​collected by the j-th current sensor at time k and time k-1, respectively, where j = 1, 2, and 3;

[0028] The anomaly detection process is as follows: If the current sensor fault diagnosis variable is less than... , If the error coefficient is greater than or equal to the error coefficient, it indicates that the current sensor may be malfunctioning. If so, it is determined that the current sensor is not malfunctioning.

[0029] Furthermore, in step 5, the phase current fault diagnosis variables are set as follows:

[0030]

[0031] In the formula, and Let A, B, and C represent the phase current values ​​of phase p at time k and time k-1, respectively.

[0032] The anomaly detection process is as follows: When the phase current sensor fault diagnosis variable is less than... , If the error coefficient is greater than or equal to the current sensor's fault diagnosis variable, then the current sensor is considered to be malfunctioning. If so, it is determined that the current sensor is not malfunctioning.

[0033] Furthermore, in step 7, the fault diagnosis variable of the switching transistor branch current... Set as actual branch current fault diagnosis variable And theoretical branch current fault diagnosis variables The difference:

[0034]

[0035]

[0036]

[0037] In the formula, and These represent the switching signals of the nth switch in phase p at time k and time k-1, respectively. and Let K and K-1 represent the phase current values ​​of phase p at time k and k-1, respectively. and Let p = A, B, C, n = 1, 2, j = 1, 2, 3, and take the following values:

[0038]

[0039] The fault diagnosis process is as follows: when the fault diagnosis variable of the switching transistor branch current is less than... , If the error coefficient is greater than the fault diagnostic variable for the switching transistor, then a short-circuit fault in the switching transistor branch current is determined. If the fault diagnosis variable for the switching transistor branch current is greater than or equal to the fault value, then an open-circuit fault has occurred. and less than or equal to If so, it is determined that no fault has occurred in the switching transistor branch current.

[0040] The advantages of this invention are:

[0041] This invention proposes a multi-fault diagnosis method for switched reluctance motor drive systems based on current behavior. Using the motor's inherent three current sensors without increasing the number of sensors, it achieves real-time monitoring of the health status of each sensor, multiple branches, and motor windings by simply changing the measurement path and the number of current measurement branches for each sensor. This method can diagnose and locate various fault types, including current sensor faults, winding open-circuit faults, power diode open-circuit faults, and mixed open-circuit and short-circuit faults in the switching transistors. Therefore, this invention requires no additional hardware and does not rely on control methods. It can quickly determine the health status of the motor drive system simply by measuring and calculating current behavior in real time. Furthermore, it has strong diagnostic capabilities, simultaneously diagnosing 12 faults across 6 fault types, achieving rapid fault diagnosis and location, and significantly improving diagnostic accuracy and speed. Attached Figure Description

[0042] The above and / or other features and advantages of the present invention will become more readily understood from the following description with reference to the accompanying drawings, in which:

[0043] Figure 1 This is an installation diagram of the current sensor in the current reconstruction technology of this invention;

[0044] Figure 2 This is a flowchart of the fault diagnosis method proposed in this invention;

[0045] Figure 3 The current reconstruction experimental results are shown in two steady-state conditions: (a) is the experimental waveform under 300 r / min and no load, and (b) is the experimental waveform under 500 r / min and 1 Nm.

[0046] Figure 4The current reconstruction experimental results are shown in two transient conditions: (a) is the experimental waveform under the conditions of sudden speed change, 300-800 r / min, and no load; (b) is the experimental waveform under the conditions of sudden load change, 500 r / min, and 0-0.8 Nm.

[0047] Figure 5 The results of the single open-circuit fault diagnosis experiment for switch S1 are shown in (a) and (b) respectively. (a) shows the experimental waveform at 300 r / min and 0.2 Nm, and (b) shows the experimental waveform at 800 r / min and 1 Nm.

[0048] Figure 6 The results of the fault diagnosis experiment for double open circuit of switching transistors S1 and S2 are shown in (a) and (b) respectively. (a) shows the experimental waveform at 300 r / min and 0.2 Nm, and (b) shows the experimental waveform at 800 r / min and 1 Nm.

[0049] Figure 7 The test results for single short-circuit fault diagnosis of switch S1 are shown in (a) and (b) respectively. (a) shows the experimental waveform at 300 r / min and 0.2 Nm, and (b) shows the experimental waveform at 800 r / min and 1 Nm.

[0050] Figure 8 The results of the test for diagnosing double short-circuit faults of switching transistors S1 and S2 are shown in (a) and (b) respectively. The waveforms are at 300 r / min and 0.2 Nm.

[0051] Figure 9 Results of the open-circuit and short-circuit mixed fault diagnosis experiment;

[0052] Figure 10 Results of the winding open-circuit fault diagnosis experiment;

[0053] Figure 11 Results of diode open-circuit fault diagnosis experiment;

[0054] Figure 12 The results are from an experiment on fault diagnosis of a current sensor. Detailed Implementation

[0055] The present invention will now be described in detail with reference to the accompanying drawings and exemplary embodiments thereof. It should be noted that the following detailed description of the present invention is for illustrative purposes only and is not intended to limit the scope of the invention.

[0056] This invention addresses the problems in fault diagnosis of switched reluctance motor (SRM) drive systems, such as the unmeasurability of some component faults, dependence on high-cost sensors, and poor real-time performance and sensitivity. It proposes a multi-fault diagnosis method for SRM drive systems based on current behavior, applicable to asymmetrical half-bridge power converters. This method, using only the motor's inherent three current sensors without increasing the number of sensors, precisely reconstructs the three-phase current values ​​by simply reselecting the sensor installation locations, changing the measurement path and the number of current measurement branches for each sensor, and using this method to monitor the health status of various sensors, multiple branches, and motor windings within the SRM drive system in real time. After a fault occurs, the monitored current value of the corresponding component deviates significantly from the theoretical current value. By analyzing this current deviation behavior, various component faults in the SRM drive system can be diagnosed and accurately located.

[0057] Reference Figure 1 In the multi-fault diagnosis method for switched reluctance motor drive systems based on current behavior provided by this invention, the first step S1 is to install the first sensor, the second sensor, and the third sensor (corresponding to the method shown in the figure) in the switched reluctance motor drive system. Figure 1 The current sensors “CURRENT SENSOR” 1, 2 and 3 are included.

[0058] like Figure 1 As shown, the drive circuits for each phase of the switched reluctance motor consist of winding branches L A L B L C The H-bridge consists of two half-bridge power device branches on both sides. Each phase has two switching transistors and two power diodes on both sides to form an H-bridge. The switching transistors are S1, S2, S3, S4, S5 and S6, and the power diodes are D1, D2, D3, D4, D5 and D6. Each phase includes a left-side switching transistor branch xLH, a left-side power diode branch xLL, a right-side switching transistor branch xRL and a right-side power diode branch xRH, where x represents phase A, B or C.

[0059] The first current sensor is connected across the common node of the C-phase right-side switch branch CRL, the A-phase winding branch LA, and the B-phase left-side switch branch BLH, and is used to monitor their combined current. The second current sensor is connected across the common node of the A-phase right-side switch branch ARL, the B-phase winding branch LB, and the C-phase left-side switch branch CLH, and is used to monitor their combined current. The third current sensor is connected across the common node of the B-phase right-side switch branch BRL, the C-phase winding branch LC, and the A-phase left-side switch branch ALH, and is used to monitor their combined current.

[0060] Reference Figure 2Next, fault diagnosis will begin. First, in step S2, the current values ​​measured by the three current sensors will be collected. Based on the current sensor installation method in step S1, the current equations measured by the three current sensors can be established as follows:

[0061] (1)

[0062] In the formula, , , i1, i2, and i3 are the phase currents of the three-phase windings, respectively, and the measured values ​​of the first, second, and third current sensors are the measured values ​​of the third current sensor, respectively. , , , , , The on / off states of switching transistors S1, S2, S3, S4, S5, and S6 are shown in the following formulas:

[0063] (2)

[0064] Where m= , , , , , .

[0065] Therefore, reconstructing the three-phase currents yields the following solution:

[0066] (3)

[0067] Then, in step S3, the current sensor fault diagnosis variables are set according to the collected current sensor current values:

[0068] (4)

[0069] in, Let be the current value at time k collected by the j-th (j=1, 2, 3) current sensor. Let be the current value collected by the j-th current sensor at time k-1.

[0070] When the current sensor fault diagnosis variable is less than ,in An error coefficient set to account for environmental influences such as sampling errors, with a value of [0, 1], indicates that the current sensor may be malfunctioning, and step S4 is performed. If the current sensor fault diagnosis variable is greater than or equal to... If the current sensor is found to be functioning correctly, proceed to step S5.

[0071] Step S4 is: continuously monitor the current sensor fault diagnosis variable at two time points. If both values ​​are close to zero, it is determined that a current sensor fault has occurred, and proceed to step S8. If no value close to zero is continuously observed, it is determined that no current sensor fault has occurred, and proceed to step S5.

[0072] Step S5 is: Set the phase current fault diagnosis variables according to the solved winding phase current values:

[0073] (5)

[0074] in and These represent the phase current values ​​of phase p (p=A, B, C) at time k and time k-1, respectively.

[0075] When the phase current fault diagnosis variable is less than ,in An error coefficient set to account for environmental influences such as sampling errors, with a value of [0, 1], indicates that the current sensor is malfunctioning, and step S6 is performed. If the phase current fault diagnosis variable is greater than or equal to... If the current sensor is found to be functioning correctly, proceed to step S7.

[0076] Step S6 is: Determine whether the phase current fault diagnosis variable at the next moment is normal. If it is normal, it is determined that a diode open circuit fault has occurred, and proceed to step S8. If it is still abnormal, it is determined that a winding open circuit fault has occurred, and also proceed to step S8.

[0077] Step S7 is as follows: Based on the collected current sensor current value and the solved winding phase current value, set the fault diagnosis variables for each switching transistor, i.e., the branch current fault diagnosis variables. It consists of two parts: theoretical branch current fault diagnosis variables. And actual branch current fault diagnosis variables :

[0078] (6)

[0079] (7)

[0080] in and Let these represent the switching signals of the nth (n=1,2) switch transistor in phase p (p=A, B, C) at time k and k-1, respectively. and Let be the current values ​​collected by the j-th current sensor at time k and time k-1, respectively. The formula is... The value of j in the table below is taken:

[0081]

[0082] The branch current fault diagnosis variables are:

[0083] (8)

[0084] The presence of a fault can be determined by the difference between the actual branch current fault diagnosis variable and the theoretical branch current fault diagnosis variable. Less than ,in An error coefficient set to account for environmental influences such as sampling errors, with a value of [0, 1], indicates that a short-circuit fault has occurred in the switching transistor. Greater than It can be determined that an open circuit fault has occurred in the switching transistor, and the process proceeds to step S8. Greater than or equal to and less than or equal to If no fault occurs in the switching transistor branch current, return to step 2 to continue fault diagnosis.

[0085] Finally, in step S8, the fault diagnosis results are output, that is, all fault types and fault points are output.

[0086] Therefore, as described above, this invention requires no additional hardware. It only uses the three inherent current sensors of the motor and does not rely on control methods. By simply changing the measurement path of each sensor and the number of current measurement branches, it measures and calculates the current behavior in real time, thereby achieving real-time monitoring of the health status of each sensor, multiple branches, and motor windings in the drive system. It can quickly determine the health status of the motor drive system within only one control cycle (setting fault diagnosis variables for time k and k-1 in the case of switch tube faults reflects one control cycle), two control cycles (the next moment in step S6 diode open circuit fault diagnosis and winding fault detection reflects two control cycles), and four control cycles (current sensor fault detection reflects four control cycles). Moreover, it has strong diagnostic capabilities and can simultaneously diagnose 12 types of faults across 6 fault categories, namely current sensor faults, winding open circuit faults, power diode open circuit faults, and mixed open-circuit and short-circuit faults of switch tubes. This enables rapid fault diagnosis and location of the switched reluctance motor drive system, greatly improving diagnostic accuracy and speed.

[0087] The following examples further illustrate the multi-fault diagnosis method for switched reluctance motor drive systems based on current behavior provided by this invention, and the proposed method is verified through a large number of experiments.

[0088] In this example, after installing three current sensors as described above, fault diagnosis of the switched reluctance motor drive system was initiated.

[0089] Figure 3 In the figure, (a) shows the waveforms of the actual current and the reconstructed current under no-load conditions of 300 r / min, and (b) shows the actual current and the reconstructed current under load conditions of 800 r / min and 1 Nm. It can be seen that the reconstructed current and the actual current can match well, which verifies the accuracy of the current reconstruction method proposed in this invention.

[0090] Figure 4 In the figure, (a) shows the actual current and reconstructed current waveforms under the transient speed condition, and (b) shows the actual current and reconstructed current waveforms under the transient load condition. It can be seen that the reconstructed current is consistent with the actual phase current under both the transient speed condition and the transient load condition, which verifies that the proposed current reconstruction method has good robustness.

[0091] Figure 5 Figures (a) and (b) show that an open circuit fault suddenly occurs at 300 r / min with a load of 0.2 Nm and 800 r / min with a load of 1 Nm, respectively. The theoretical current diagnostic variable and the actual current diagnostic variable rapidly become abnormal after the fault occurs. At the same time, the theoretical diagnostic variable is less than the actual diagnostic variable. Therefore, an open circuit fault is diagnosed and the location of the fault is determined.

[0092] Figure 6 (a) shows the experimental waveform at 300 r / m and 0.2 Nm. After two switching transistors experience open-circuit faults, the fault diagnosis variables of the corresponding devices will differ significantly from the theoretical variables, thus enabling fault diagnosis. Similarly, in Figure 6 (b) shows that the two switching transistors can also be used for diagnosis when the load is 800 r / min and 1 Nm.

[0093] Figure 7 In Figures (a) and (b), an open-circuit fault suddenly occurs under the conditions of 300 r / min load of 0.2 Nm and 800 r / min load of 1 Nm, respectively. The theoretical current diagnostic variable and the actual current diagnostic variable rapidly become abnormal after the fault occurs. At the same time, the theoretical diagnostic variable is greater than the actual diagnostic variable. Therefore, a short-circuit fault is diagnosed and the location of the fault is determined.

[0094] Figure 8(a) shows the experimental waveform at 300 r / m and 0.2 Nm. After two switching transistors experience open-circuit faults, the fault diagnosis variables of the corresponding devices will differ significantly from the theoretical variables, thus enabling fault diagnosis. Similarly, in Figure 8 When the load condition of 800r / min1Nm is shown in (b), the fault diagnosis function of the two switching transistors can also be realized.

[0095] Figure 9 In this setup, the S1 transistor in phase A is assumed to have an open-circuit fault, and the S2 transistor to have a short-circuit fault. This is verified through testing at 800 rpm and a 1 Nm load. Figure 9 As shown, when an open circuit fault occurs in S1, the corresponding S1 diagnostic variable will show an open circuit fault anomaly, and similarly, the S2 diagnostic variable will show a short circuit fault anomaly.

[0096] Depend on Figure 10 As can be seen, after an open circuit occurs in phase A, the current value will remain at 0. To distinguish between an open circuit fault in the winding and an open circuit fault in the diode, all the switching devices in the asymmetrical half-bridge are turned on when the phase current becomes zero. However, the phase current does not increase, indicating that an open circuit fault has occurred in the winding.

[0097] Depend on Figure 11 It can be seen that after a diode open circuit fault occurs in D1, the phase current will drop rapidly to 0 within one control cycle. After one control cycle, the current increases, indicating that the fault is not a winding open circuit fault. After two consecutive control cycles, S1 is turned off, and the current becomes 0 again at the next moment, indicating that the fault point is in D1. Thus, the diagnosis and location of the diode open circuit fault are realized.

[0098] Figure 12 In this example, a sudden current sensor fault is simulated under no-load conditions at 500 rpm. Upon the fault occurring, the current changes from its normal value to zero. Therefore, the fault diagnosis variable set for the current sensor will become zero at the next moment, thus achieving the diagnosis and location of the current sensor fault. This example verifies that the method proposed in this invention can quickly determine the health status of the motor drive system within only 1, 2, or 4 control cycles, and can simultaneously and accurately diagnose 12 faults across 6 fault types, greatly improving the accuracy and speed of fault diagnosis.

[0099] Finally, it should be noted that the features mentioned and / or shown in the above description of exemplary embodiments of the present invention can be combined in the same or similar manner with one or more other embodiments, combined with features in other embodiments, or substituted for corresponding features in other embodiments. These combined or substituted technical solutions should also be considered to be included within the scope of protection of the present invention.

Claims

1. A multi-fault diagnosis method for a switched reluctance motor drive system based on current behavior, characterized in that, Includes the following steps: Step 1: Install a first current sensor, a second current sensor, and a third current sensor in the switched reluctance motor drive system; wherein, each phase drive circuit of the switched reluctance motor consists of winding branches LA, LB, LC, and power device branches on both sides of the half-bridge. The two switching transistors and two power diodes on both sides of each phase form an H-bridge. Each phase includes a left-side switching transistor branch xLH, a left-side power diode branch xLL, a right-side switching transistor branch xRL, and a right-side power diode branch xRH, where x represents phase A, B, or C; The first current sensor is connected across the common node of the C-phase right-side switch branch CRL, the A-phase winding branch LA, and the B-phase left-side switch branch BLH, and is used to monitor their combined current; the second current sensor is connected across the common node of the A-phase right-side switch branch ARL, the B-phase winding branch LB, and the C-phase left-side switch branch CLH, and is used to monitor their combined current; the third current sensor is connected across the common node of the B-phase right-side switch branch BRL, the C-phase winding branch LC, and the A-phase left-side switch branch ALH, and is used to monitor their combined current. Step 2: Collect the current values ​​measured by the first current sensor, the second current sensor, and the third current sensor, and calculate the winding phase current value based on the collected current sensor current values; Step 3: Set the current sensor fault diagnosis variable according to the collected current sensor current value, and determine whether the first current sensor, the second current sensor and the third current sensor are abnormal according to the current sensor fault diagnosis variable. If it is determined that there may be an abnormality, proceed to step 4; if it is determined that there is no abnormality, proceed to step 5. Step 4: Continuously monitor the current sensor fault diagnosis variables at two different times. If both values ​​are close to zero, it is determined that a current sensor fault has occurred, and proceed to step 8. If no values ​​close to zero are continuously observed, it is determined that no current sensor fault has occurred, and proceed to step 5. Step 5: Set phase current fault diagnosis variables according to the solved winding phase current values, and determine whether the first current sensor, the second current sensor and the third current sensor are abnormal according to the phase current fault diagnosis variables. If an abnormality is found, proceed to step 6; if no abnormality is found, proceed to step 7. Step 6: Determine whether the phase current fault diagnosis variable is normal at the next moment. If it is normal, it is determined that a diode open circuit fault has occurred, and proceed to step 8. If it is still abnormal, it is determined that a winding open circuit fault has occurred, and proceed to step 8. Step 7: Set the switch tube branch current fault diagnosis variable according to the collected current sensor current value and the solved winding phase current value, and determine whether a switch tube branch current fault has occurred based on the switch tube branch current fault diagnosis variable. If a switch tube open circuit or short circuit fault occurs, proceed to step 8. If no fault occurs, return to step 2 to continue fault diagnosis. Step 8: Output the fault diagnosis results.

2. The multi-fault diagnosis method for a switched reluctance motor drive system based on current behavior according to claim 1, characterized in that, In step 2, the winding phase current is solved according to the following sub-steps: Step 2.1: Establish the current equations measured by the first current sensor, the second current sensor, and the third current sensor: In the formula, , , i1, i2, and i3 are the phase currents of the three-phase windings, respectively, and are the measured values ​​of the first current sensor, the second current sensor, and the third current sensor, respectively. , , , , , These represent the on / off status of switching transistors S1, S2, S3, S4, S5, and S6, respectively. Step 2.2: Solve for the winding phase currents based on the established current equations: 。 3. The multi-fault diagnosis method for a switched reluctance motor drive system based on current behavior according to claim 2, characterized in that, , , , , , The value is determined as follows: In the formula, m= , , , , , .

4. The multi-fault diagnosis method for a switched reluctance motor drive system based on current behavior according to claim 1 or 2, characterized in that, In step 3, the current sensor fault diagnosis variables are set as follows: In the formula, and These are the current values ​​collected by the j-th current sensor at time k and time k-1, respectively, where j = 1, 2, and 3; The anomaly detection process is as follows: If the current sensor fault diagnosis variable is less than... , If the error coefficient is greater than or equal to the error coefficient, it indicates that the current sensor may be malfunctioning. If so, it is determined that the current sensor is not malfunctioning.

5. The multi-fault diagnosis method for a switched reluctance motor drive system based on current behavior according to claim 1 or 2, characterized in that, In step 5, the phase current fault diagnosis variables are set as follows: In the formula, and Let A, B, and C represent the phase current values ​​of phase p at time k and time k-1, respectively. The anomaly detection process is as follows: When the phase current sensor fault diagnosis variable is less than... , If the error coefficient is greater than or equal to the current sensor's fault diagnosis variable, then the current sensor is considered to be malfunctioning. If so, it is determined that the current sensor is not malfunctioning.

6. The multi-fault diagnosis method for a switched reluctance motor drive system based on current behavior according to claim 1 or 2, characterized in that, In step 7, the fault diagnosis variable of the switching transistor branch current Set as actual branch current fault diagnosis variable And theoretical branch current fault diagnosis variables The difference: In the formula, and These represent the switching signals of the nth switch in phase p at time k and time k-1, respectively. and Let K and K-1 represent the phase current values ​​of phase p at time k and k-1, respectively. and Let p = A, B, C, n = 1, 2, j = 1, 2, 3, and their values ​​be as follows: The fault diagnosis process is as follows: when the fault diagnosis variable of the switching transistor branch current is less than... , If the error coefficient is greater than the fault diagnostic variable for the switching transistor, then a short-circuit fault in the switching transistor branch current is determined. If the fault diagnosis variable for the switching transistor branch current is greater than or equal to the fault value, then an open-circuit fault has occurred. and less than or equal to If so, it is determined that no fault has occurred in the switching transistor branch current.

Images