RC-IGBT fault detection and diagnosis method

Abstract

The invention discloses an RC-IGBT fault detection and diagnosis method, and belongs to the technical field of IGBT fault detection. Pulse signals with different frequencies (low frequency, intermediate frequency and high frequency) are applied to an RC-IGBT tube, a voltage signal between a collector electrode and an emitter electrode is analyzed, a distortion value sequence is constructed, and the RC-IGBT fault detection and diagnosis method is realized. And the RC-IGBT fault diagnosis value is obtained by combining the distortion time deterioration coefficient and the distortion frequency deterioration coefficient under the low frequency, the intermediate frequency and the high frequency, so that the RC-IGBT fault diagnosis precision is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of IGBT fault detection, and in particular to a RC-IGBT fault detection and diagnosis method. Background Art

[0002] RC-IGBT (reverse conducting insulated gate bipolar transistor) is a key power semiconductor device in modern power electronic systems and is widely used in high-voltage converters, motor drives, and new energy power systems. Traditional IGBT fault detection methods mainly rely on static parameter measurements and simple electrical characteristics analysis, which have many limitations. Existing technologies usually use single parameter monitoring, such as DC parameter measurement, switching loss analysis, or temperature characteristics detection. These methods are difficult to accurately reflect the actual working status and potential faults of RC-IGBT, as well as its response to different switching frequencies.

[0003] Traditional methods mainly rely on a single parameter, such as static parameter measurement (such as drain-source saturation voltage VCE(sat)) or DC parameter analysis, which cannot fully capture the dynamic characteristic changes of the device. When RC-IGBT operates in a complex working environment, a single parameter is difficult to accurately reflect its actual performance degradation. Therefore, the existing technology has the problem of low accuracy of RC-IGBT fault diagnosis. Summary of the invention

[0004] In view of the above-mentioned deficiencies in the prior art, the present invention provides a RC-IGBT fault detection and diagnosis method which solves the problem of low RC-IGBT fault diagnosis accuracy in the prior art.

[0005] In order to achieve the above-mentioned invention object, the technical solution adopted by the present invention is: a RC-IGBT fault detection and diagnosis method, comprising the following steps: S1, apply low frequency, medium frequency and high frequency pulse signals to the gate of RC-IGBT tube respectively, and collect the voltage signal between the collector and emitter of RC-IGBT tube; S2, calculating the distortion value for each switching cycle in the voltage signal and constructing a distortion value sequence; S3, calculating the distortion time deterioration coefficient for the distortion value sequence; S4, taking the average of each distortion value sequence to obtain the distortion mean, and calculating the distortion frequency deterioration coefficient; S5. According to the distortion mean values ​​corresponding to low frequency, medium frequency and high frequency, based on the correction of distortion time deterioration coefficient and distortion frequency deterioration coefficient, the RC-IGBT fault diagnosis value is obtained.

[0006] Furthermore, in S1, the frequency range of the low frequency is 0.1kHz~1kHz, the frequency range of the medium frequency is 1kHz~20kHz, and the frequency range of the high frequency is 20kHz~100kHz.

[0007] Further, S2 includes the following sub-steps: S21, extracting a high-level vibration value, a low-level vibration value, a rising edge steepness value, and a falling edge steepness value according to each switching cycle in the voltage signal; S22, calculating the distortion value according to the high-level vibration value, the low-level vibration value, the rising edge steepness value and the falling edge steepness value; S23. For the same voltage signal, the distortion values ​​are used as elements in the order of occurrence in time to construct a distortion value sequence.

[0008] Furthermore, the calculation formula of the high-level vibration value in S21 is: , where V H is the high level vibration value, H i is the ith voltage value in the high level part of the switching cycle, H steady is the high voltage steady state value, i is a positive integer, N H is the number of voltage values ​​in the high level part of the switching cycle, | | is the absolute value operation; The calculation formula for the low-level vibration value in S21 is: , where V L is the low level vibration value, L i is the i-th voltage value in the low level part of the switching cycle, L steady is the low level steady state value, H L is the number of voltage values ​​in the low-level part of the switching cycle; The calculation formula of the rising edge steepness value in S21 is: , where K R is the rising edge steepness value, V R,E is the end voltage value of the rising edge in the switching cycle, V R,S is the starting voltage value of the rising edge in the switching cycle, △t R It is the interval time from the starting point to the end point of the rising edge; The calculation formula of the falling edge steepness in S21 is: , where K F is the falling edge steepness value, V F,S is the starting voltage value of the falling edge in the switching cycle, V F,E is the end voltage value of the falling edge in the switching cycle, △t F It is the interval time from the start point to the end point of the falling edge.

[0009] Furthermore, the formula for calculating the distortion value in S22 is: , where θ is the distortion value, V H is the high level vibration value, V L is the low level vibration value, K R is the rising edge steepness value, K F K is the falling edge steepness value, R,th is the steepness standard value of the rising edge, K F,th This is the standard value for the steepness of the falling edge.

[0010] Furthermore, S3 includes the following sub-steps: S31, taking the last M consecutive distortion values ​​and the first M consecutive distortion values ​​in the distortion value sequence, where M is a positive integer; S32, taking the average of the M consecutive distortion values ​​at the end to obtain the end distortion mean; S33, taking an average of the initial M continuous distortion values ​​to obtain an initial distortion average; S34. The ratio of the final distortion mean to the initial distortion mean is used as the distortion time deterioration coefficient.

[0011] Further, S4 includes the following sub-steps: S41, taking the mean of each distortion value sequence to obtain a distortion mean; S42, calculating a first distortion frequency deterioration value according to a difference between a distortion mean value corresponding to the intermediate frequency and a distortion mean value corresponding to the low frequency; S43, calculating a second distortion frequency degradation value according to a difference between a distortion mean value corresponding to the high frequency and a distortion mean value corresponding to the intermediate frequency; S44. Add the first distortion frequency degradation value and the second distortion frequency degradation value to obtain a distortion frequency degradation coefficient.

[0012] Furthermore, the formula for calculating the first distortion frequency degradation value in S42 is: , where μ1 is the first distortion frequency deterioration value, tanh is the hyperbolic tangent function, θ M is the distortion mean corresponding to the intermediate frequency, θ L is the distortion mean corresponding to low frequency; The formula for calculating the second distortion frequency degradation value in S43 is: , where μ2 is the second distortion frequency degradation value, θ H is the distortion mean corresponding to high frequency, θ M is the distortion mean corresponding to the intermediate frequency.

[0013] Further, S5 includes the following sub-steps: S51, multiplying the distortion mean value at low frequency and the distortion time deterioration coefficient to obtain a fault value at low frequency; S52, multiplying the distortion mean value at the intermediate frequency by the distortion time deterioration coefficient to obtain a fault value at the intermediate frequency; S53, multiplying the distortion mean value at high frequency by the distortion time deterioration coefficient to obtain a fault value at high frequency; S54, calculating the RC-IGBT fault diagnosis value according to the fault value at low frequency, the fault value at medium frequency and the fault value at high frequency based on the correction of the distortion frequency deterioration coefficient.

[0014] Furthermore, the formula for calculating the RC-IGBT fault diagnosis value in S54 is: , where G is the RC-IGBT fault diagnosis value, y L is the fault value at low frequency, y M is the fault value at medium frequency, y H is the fault value at high frequency, and γ is the distortion frequency deterioration coefficient.

[0015] In summary, the beneficial effects of the present invention are: 1. The present invention applies three kinds of pulse signals, low frequency, medium frequency and high frequency, to the RC-IGBT tube to comprehensively capture the voltage signal characteristics of the device under different frequency working conditions. Compared with the traditional single parameter detection method, the present invention can more accurately reflect the dynamic characteristic changes of the RC-IGBT and significantly improve the accuracy and reliability of fault detection.

[0016] 2. The present invention introduces distortion value sequence, distortion time deterioration coefficient and distortion frequency deterioration coefficient to realize comprehensive performance evaluation of RC-IGBT at low, medium and high frequencies. Through comprehensive analysis of multi-frequency signals, the actual working state of the device can be more comprehensively characterized, overcoming the limitation of traditional methods that only rely on a single frequency or a single parameter.

[0017] 3. The present invention calculates the distortion value in each switching cycle of the voltage signal to reflect the distortion of a switching cycle signal. The constructed distortion value sequence reflects the distortion of each switching cycle signal in the voltage signal, and calculates the distortion time deterioration coefficient of the distortion value sequence to reflect the distortion deterioration of the RC-IGBT tube with time. The distortion frequency deterioration coefficient is then calculated to reflect the distortion deterioration of the RC-IGBT tube with frequency. According to the distortion average at low frequency, medium frequency and high frequency, and based on the correction of the distortion time deterioration coefficient and the distortion frequency deterioration coefficient, the accuracy of RC-IGBT fault diagnosis is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a flow chart of a RC-IGBT fault detection and diagnosis method; Figure 2 It is a signal schematic diagram of a switching cycle in the desired voltage signal; Figure 3It is a signal diagram of a switching cycle in the voltage signal when the RC-IGBT fails. DETAILED DESCRIPTION

[0019] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0020] like Figure 1 As shown, a RC-IGBT fault detection and diagnosis method includes the following steps: S1, apply low frequency, medium frequency and high frequency pulse signals to the gate of RC-IGBT tube respectively, and collect the voltage signal between the collector and emitter of RC-IGBT tube; S2, calculating the distortion value for each switching cycle in the voltage signal and constructing a distortion value sequence; S3, calculating the distortion time deterioration coefficient for the distortion value sequence; S4, taking the average of each distortion value sequence to obtain the distortion mean, and calculating the distortion frequency deterioration coefficient; S5. According to the distortion mean values ​​corresponding to low frequency, medium frequency and high frequency, based on the correction of distortion time deterioration coefficient and distortion frequency deterioration coefficient, the RC-IGBT fault diagnosis value is obtained.

[0021] In this embodiment, the frequency range of the low frequency in S1 is 0.1kHz~1kHz, the frequency range of the medium frequency is 1kHz~20kHz, and the frequency range of the high frequency is 20kHz~100kHz.

[0022] The low frequency band (0.1~1kHz) represents the basic operating frequency of RC-IGBT in industrial control and power systems. It reflects the static characteristics and basic switching performance of the device. It can capture the most basic performance degradation signals of the device and simulate low-speed conversion and load regulation conditions.

[0023] The medium frequency band (1~20kHz) corresponds to the actual operating frequency of most industrial frequency conversion and motor drive systems. It is in the key conversion range of switching loss and dynamic performance. It can reflect the performance changes of RC-IGBT under complex working conditions and simulate medium load and medium speed operating conditions.

[0024] The high frequency band (20~100kHz) represents the high-speed switching characteristics of modern power electronic systems, reflects the dynamic response of RC-IGBT under extreme working conditions, captures sensitive signals of microscopic defects and performance degradation inside the device, and simulates high-frequency application scenarios such as new energy and high-speed rail traction.

[0025] In the present invention, a pulse signal is applied to the gate of the RC-IGBT tube, so that the RC-IGBT tube is turned on at a high level and turned off at a low level, thereby testing the switching performance of the RC-IGBT tube.

[0026] In this embodiment, S2 includes the following sub-steps: S21, extracting a high-level vibration value, a low-level vibration value, a rising edge steepness value, and a falling edge steepness value according to each switching cycle in the voltage signal; S22, calculating the distortion value according to the high-level vibration value, the low-level vibration value, the rising edge steepness value and the falling edge steepness value; S23. For the same voltage signal, the distortion values ​​are used as elements in the order of occurrence in time to construct a distortion value sequence.

[0027] The high-level vibration value reflects the stability of the switching device in the high-level state, and reveals the influence of parasitic parameters such as parasitic inductance and capacitance; the low-level vibration value characterizes the electrical characteristics of the device in the on state, reflects the leakage current and switching loss, and reveals the degree of degradation of the PN junction and the gate oxide layer; the falling edge steepness value reflects the turn-on speed and switching performance of the device, and directly reflects the carrier injection and transport capabilities; the rising edge steepness value reflects the turn-off speed and switching performance of the device, and reveals the carrier recombination and expulsion characteristics.

[0028] In a switching cycle of the voltage signal, the high-level vibration value, the low-level vibration value, the rising edge steepness value and the falling edge steepness value are extracted, and the distortion value is calculated, which can reflect the distortion of the signal in a switching cycle.

[0029] In this embodiment, the calculation formula of the high-level vibration value in S21 is: , where V H is the high level vibration value, H i is the ith voltage value in the high level part of the switching cycle, H steady is the high voltage steady state value, i is a positive integer, N H is the number of voltage values ​​in the high level part of the switching cycle, and | | is the absolute value operation.

[0030] In this embodiment, the calculation formula of the low-level vibration value in S21 is: , where V L is the low level vibration value, L i is the i-th voltage value in the low level part of the switching cycle, L steady is the low level steady state value, H L is the number of voltage values ​​in the low-level portion of the switching cycle.

[0031] like Figure 2As shown in the figure, when the RC-IGBT tube is normal, the high level part and the low level part are stable. Figure 3 As shown, in one switching cycle, there are a large number of oscillation signals in the high level part and the low level part, which seriously deviate from the low level steady-state value and the high level steady-state value.

[0032] In this embodiment, the high voltage steady-state value is the bus voltage of the RC-IGBT tube, and the low voltage steady-state value is the ground voltage.

[0033] In this embodiment, the calculation formula of the rising edge steepness value in S21 is: , where K R is the rising edge steepness value, V R,E is the end voltage value of the rising edge in the switching cycle, V R,S is the starting voltage value of the rising edge in the switching cycle, △t R It is the interval time from the starting point to the end point of the rising edge.

[0034] The end voltage value of the rising edge is the last voltage value in the rising edge segment signal, and the start voltage value of the rising edge is the first voltage value in the rising edge segment signal.

[0035] In this embodiment, the calculation formula of the falling edge steepness value in S21 is: , where K F is the falling edge steepness value, V F,S is the starting voltage value of the falling edge in the switching cycle, V F,E is the end voltage value of the falling edge in the switching cycle, △t F It is the interval time from the start point to the end point of the falling edge.

[0036] The starting voltage value of the falling edge is the first voltage value in the falling edge signal, and the ending voltage value of the falling edge is the last voltage value in the falling edge segment signal.

[0037] The steepness of the rising edge and the falling edge of the present invention directly reflects the switching speed and stability of the IGBT. By monitoring the changes of these values, potential faults or performance degradation can be discovered in time, thereby improving the accuracy of fault diagnosis.

[0038] In this embodiment, the formula for calculating the distortion value in S22 is: , where θ is the distortion value, V H is the high level vibration value, V L is the low level vibration value, K R is the rising edge steepness value, K F K is the falling edge steepness value, R,th is the steepness standard value of the rising edge, K F,th is the steepness standard value of the falling edge, and || is the absolute value operation.

[0039] In the present invention, the higher the high-level vibration value and the low-level vibration value are, and the more the rising edge steepness value deviates from the rising edge steepness standard value, and the more the falling edge steepness value deviates from the falling edge steepness standard value, the greater the distortion value is.

[0040] The steep standard value of the rising edge and the steep standard value of the falling edge can be found in the technical manual of the RC-IGBT tube. For a typical RC-IGBT tube, the specific values ​​of the steep standard value of the rising edge and the steep standard value of the falling edge are usually in the following ranges: steep standard value of the rising edge: 5000-15000 V / μs, high-performance IGBT: up to 10-20kV / μs, low-power IGBT: about 2-5kV / μs, steep standard value of the falling edge: 3000-10000V / μs, high-performance IGBT: up to 8-15kV / μs, low-power IGBT: about 1-4kV / μs.

[0041] In this embodiment, S3 includes the following sub-steps: S31, taking the last M consecutive distortion values ​​and the first M consecutive distortion values ​​in the distortion value sequence, where M is a positive integer; S32, taking the average of the M consecutive distortion values ​​at the end to obtain the end distortion mean; S33, taking an average of the initial M continuous distortion values ​​to obtain an initial distortion average; S34. The ratio of the final distortion mean to the initial distortion mean is used as the distortion time deterioration coefficient.

[0042] The present invention compares the final distortion value with the initial distortion value to reflect whether the distortion situation worsens over time.

[0043] In this embodiment, S4 includes the following sub-steps: S41, taking the mean of each distortion value sequence to obtain a distortion mean; S42, calculating a first distortion frequency deterioration value according to a difference between a distortion mean value corresponding to the intermediate frequency and a distortion mean value corresponding to the low frequency; S43, calculating a second distortion frequency degradation value according to a difference between a distortion mean value corresponding to the high frequency and a distortion mean value corresponding to the intermediate frequency; S44. Add the first distortion frequency degradation value and the second distortion frequency degradation value to obtain a distortion frequency degradation coefficient.

[0044] In this embodiment, the formula for calculating the first distortion frequency degradation value in S42 is: , where μ1 is the first distortion frequency deterioration value, tanh is the hyperbolic tangent function, θ M is the distortion mean corresponding to the intermediate frequency, θ L is the distortion mean corresponding to low frequency; The formula for calculating the second distortion frequency degradation value in S43 is: , where μ2 is the second distortion frequency degradation value, θ H is the distortion mean corresponding to high frequency, θ M is the distortion mean corresponding to the intermediate frequency.

[0045] The present invention calculates a first distortion frequency deterioration value according to a difference between a distortion mean value corresponding to an intermediate frequency and a distortion mean value corresponding to a low frequency, reflecting the deterioration of the RC-IGBT tube from a low frequency to an intermediate frequency, and then calculates a second distortion frequency deterioration value according to a difference between a distortion mean value corresponding to a high frequency and a distortion mean value corresponding to an intermediate frequency, reflecting the deterioration of the RC-IGBT tube from an intermediate frequency to a high frequency.

[0046] When calculating the distortion frequency deterioration value, the present invention takes the cube of the difference of the distortion mean value, enhances the difference situation, does not change the positive and negative of the difference, sets the hyperbolic tangent function so that the distortion frequency deterioration value is within the range of -1 to 1, when the distortion frequency deterioration value is greater than 0, the state of the RC-IGBT tube further deteriorates with the increase of frequency, and when the distortion frequency deterioration value is less than 0, the state of the RC-IGBT tube improves with the increase of frequency.

[0047] In this embodiment, S5 includes the following sub-steps: S51, multiplying the distortion mean value at low frequency and the distortion time deterioration coefficient to obtain a fault value at low frequency; S52, multiplying the distortion mean value at the intermediate frequency by the distortion time deterioration coefficient to obtain a fault value at the intermediate frequency; S53, multiplying the distortion mean value at high frequency by the distortion time deterioration coefficient to obtain a fault value at high frequency; S54, calculating the RC-IGBT fault diagnosis value according to the fault value at low frequency, the fault value at medium frequency and the fault value at high frequency based on the correction of the distortion frequency deterioration coefficient.

[0048] Since the present invention applies three pulse signals, corresponding to three voltage signals and three distortion value sequences, the distortion mean corresponding to the low frequency is multiplied by the distortion time deterioration coefficient corresponding to the low frequency to obtain the fault value at the low frequency; the distortion mean corresponding to the medium frequency is multiplied by the distortion time deterioration coefficient corresponding to the medium frequency to obtain the fault value at the medium frequency; the distortion mean corresponding to the high frequency is multiplied by the distortion time deterioration coefficient corresponding to the high frequency to obtain the fault value at the high frequency. When the distortion time deterioration coefficient is greater than 1, as the working time increases, the state of the RC-IGBT tube further deteriorates, therefore, it has an enhancing effect on the distortion mean, and the obtained fault value is larger. When the distortion time deterioration coefficient is less than 1, as the working time increases, the state of the RC-IGBT tube improves, therefore, it has a weakening effect on the distortion mean, and the obtained fault value is smaller.

[0049] In this embodiment, the formula for calculating the RC-IGBT fault diagnosis value in S54 is: , where G is the RC-IGBT fault diagnosis value, y L is the fault value at low frequency, y M is the fault value at medium frequency, y H is the fault value at high frequency, and γ is the distortion frequency deterioration coefficient.

[0050] The present invention combines the fault values ​​at low frequency, medium frequency and high frequency, and further improves the accuracy of fault diagnosis based on the compensation of the distortion frequency deterioration coefficient.

[0051] The present invention applies low-frequency, medium-frequency and high-frequency pulse signals to the RC-IGBT tube to comprehensively capture the voltage signal characteristics of the device under different frequency working conditions. Compared with the traditional single parameter detection method, the present invention can more accurately reflect the dynamic characteristic changes of the RC-IGBT and significantly improve the accuracy and reliability of fault detection.

[0052] The present invention introduces distortion value sequence, distortion time deterioration coefficient and distortion frequency deterioration coefficient to realize comprehensive performance evaluation of RC-IGBT at low, medium and high frequencies. Through comprehensive analysis of multi-frequency signals, the actual working state of the device can be more comprehensively characterized, overcoming the limitation of traditional methods that only rely on a single frequency or a single parameter.

[0053] The present invention calculates the distortion value in each switching cycle of the voltage signal to reflect the distortion of a switching cycle signal. The constructed distortion value sequence reflects the distortion of each switching cycle signal in the voltage signal. The distortion time deterioration coefficient is calculated for the distortion value sequence to reflect the distortion deterioration of the RC-IGBT tube with time. The distortion frequency deterioration coefficient is then calculated to reflect the distortion deterioration of the RC-IGBT tube with frequency. According to the distortion mean values ​​at low frequency, medium frequency and high frequency, the distortion time deterioration coefficient and the distortion frequency deterioration coefficient are corrected to improve the accuracy of RC-IGBT fault diagnosis.

[0054] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A RC-IGBT fault detection and diagnosis method, characterized in that: The following steps are involved: S1, apply low frequency, medium frequency and high frequency pulse signals to the gate of RC-IGBT tube respectively, and collect the voltage signal between the collector and emitter of RC-IGBT tube; S2, calculating the distortion value for each switching cycle in the voltage signal and constructing a distortion value sequence; S3, calculating the distortion time deterioration coefficient for the distortion value sequence; S4, taking the average of each distortion value sequence to obtain the distortion mean, and calculating the distortion frequency deterioration coefficient; S5. According to the distortion mean values ​​corresponding to low frequency, medium frequency and high frequency, based on the correction of distortion time deterioration coefficient and distortion frequency deterioration coefficient, the RC-IGBT fault diagnosis value is obtained.

2. The RC-IGBT fault detection and diagnosis method according to claim 1, characterized in that: The frequency range of the low frequency in S1 is 0.1kHz~1kHz, the frequency range of the medium frequency is 1kHz~20kHz, and the frequency range of the high frequency is 20kHz~100kHz.

3. The RC-IGBT fault detection and diagnosis method according to claim 1, characterized in that: The S2 comprises the following sub-steps: S21, extracting a high-level vibration value, a low-level vibration value, a rising edge steepness value, and a falling edge steepness value according to each switching cycle in the voltage signal; S22, calculating the distortion value according to the high-level vibration value, the low-level vibration value, the rising edge steepness value and the falling edge steepness value; S23. For the same voltage signal, the distortion values ​​are used as elements in the order of occurrence in time to construct a distortion value sequence.

4. The RC-IGBT fault detection and diagnosis method according to claim 3, characterized in that: The calculation formula of the high-level vibration value in S21 is: , where V H is the high level vibration value, H i is the ith voltage value in the high level part of the switching cycle, H steady is the high voltage steady state value, i is a positive integer, N H is the number of voltage values ​​in the high level part of the switching cycle, | | is the absolute value operation; The calculation formula of the low-level vibration value in S21 is: , where V L is the low level vibration value, L i is the i-th voltage value in the low level part of the switching cycle, L steady is the low level steady state value, H L is the number of voltage values ​​in the low-level part of the switching cycle; The calculation formula of the rising edge steepness value in S21 is: , where K R is the rising edge steepness value, V R,E is the end voltage value of the rising edge in the switching cycle, V R,S is the starting voltage value of the rising edge in the switching cycle, △t R It is the interval time from the starting point to the end point of the rising edge; The calculation formula of the falling edge steepness value in S21 is: , where K F is the falling edge steepness value, V F,S is the starting voltage value of the falling edge in the switching cycle, V F,E is the end voltage value of the falling edge in the switching cycle, △t F It is the interval time from the start point to the end point of the falling edge.

5. The RC-IGBT fault detection and diagnosis method according to claim 3, characterized in that: The formula for calculating the distortion value in S22 is: , where θ is the distortion value, V H is the high level vibration value, V L is the low level vibration value, K R is the rising edge steepness value, K F K is the falling edge steepness value, R,th is the steepness standard value of the rising edge, K F,th is the steepness standard value of the falling edge, and || is the absolute value operation.

6. The RC-IGBT fault detection and diagnosis method according to claim 1, characterized in that: The S3 comprises the following sub-steps: S31, taking the last M consecutive distortion values ​​and the first M consecutive distortion values ​​in the distortion value sequence, where M is a positive integer; S32, taking the average of the M consecutive distortion values ​​at the end to obtain the end distortion mean; S33, taking an average of the initial M continuous distortion values ​​to obtain an initial distortion average; S34. The ratio of the final distortion mean to the initial distortion mean is used as the distortion time deterioration coefficient.

7. The RC-IGBT fault detection and diagnosis method according to claim 1, characterized in that: The S4 comprises the following sub-steps: S41, taking the mean of each distortion value sequence to obtain a distortion mean; S42, calculating a first distortion frequency deterioration value according to a difference between a distortion mean value corresponding to the intermediate frequency and a distortion mean value corresponding to the low frequency; S43, calculating a second distortion frequency degradation value according to a difference between a distortion mean value corresponding to the high frequency and a distortion mean value corresponding to the intermediate frequency; S44. Add the first distortion frequency degradation value and the second distortion frequency degradation value to obtain a distortion frequency degradation coefficient.

8. The RC-IGBT fault detection and diagnosis method according to claim 7, characterized in that: The formula for calculating the first distortion frequency degradation value in S42 is: , where μ1 is the first distortion frequency deterioration value, tanh is the hyperbolic tangent function, θ M is the distortion mean corresponding to the intermediate frequency, θ L is the distortion mean corresponding to low frequency; The formula for calculating the second distortion frequency degradation value in S43 is: , where μ2 is the second distortion frequency degradation value, θ H is the distortion mean corresponding to high frequency, θ M is the distortion mean corresponding to the intermediate frequency.

9. The RC-IGBT fault detection and diagnosis method according to claim 1, characterized in that: The S5 comprises the following sub-steps: S51, multiplying the distortion mean value at low frequency and the distortion time deterioration coefficient to obtain a fault value at low frequency; S52, multiplying the distortion mean value at the intermediate frequency by the distortion time deterioration coefficient to obtain a fault value at the intermediate frequency; S53, multiplying the distortion mean value at high frequency and the distortion time deterioration coefficient to obtain a fault value at high frequency; S54, calculating the RC-IGBT fault diagnosis value according to the fault value at low frequency, the fault value at medium frequency and the fault value at high frequency based on the correction of the distortion frequency deterioration coefficient.

10. The RC-IGBT fault detection and diagnosis method according to claim 9, characterized in that: The formula for calculating the RC-IGBT fault diagnosis value in S54 is: , where G is the RC-IGBT fault diagnosis value, y L is the fault value at low frequency, y M is the fault value at medium frequency, y H is the fault value at high frequency, and γ is the distortion frequency deterioration coefficient.

Images