Combined fault diagnosis method for three-phase four-leg wind power inverter

Through the methods of non-singular matrix decoupling and reduced-order adaptive gain sliding mode observer, sensor faults and switch tube faults are decomposed into independent subsystems, which solves the problems of high fault diagnosis complexity and insufficient accuracy in the existing technology and realizes fast and accurate fault detection.

CN120779130APending Publication Date: 2025-10-14HEFEI UNIV OF TECH
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202510764278.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-06-09
Publication Date
2025-10-14

AI Technical Summary

Technical Problem

Existing fault diagnosis technology for three-phase four-leg inverters has problems such as difficult signal processing, large computational complexity, high diagnostic complexity, and insufficient model accuracy. In particular, sensor faults and switch tube faults are easily affected by each other, resulting in long diagnostic time and insufficient accuracy.

Method used

Non-singular matrix decoupling technology is used to decompose sensor fault and switch tube fault into two non-interfering subsystems, and a reduced-order adaptive gain sliding mode observer is established. By designing an adaptive detection threshold, the fault detection time is reduced and the robustness of the diagnosis is improved.

Benefits of technology

The simultaneous diagnosis of sensor faults and switch tube faults is achieved, which avoids the mutual influence between faults and improves the accuracy and robustness of fault detection. At the same time, no additional sensors are required, which reduces the cost of fault detection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120779130A_ABST
    Figure CN120779130A_ABST
Patent Text Reader

Abstract

The invention provides a combined fault diagnosis method for a three-phase four-leg wind power inverter, and belongs to the technical field of fault diagnosis of power appliances. The method comprises the steps of establishing a hybrid logic dynamic model of a three-phase four-bridge-arm wind power inverter, establishing a reduced-order adaptive gain sliding mode observer to accurately estimate a system state and a sensor fault, and performing switch tube open-circuit fault diagnosis through a residual error and a norm threshold value of the observer. And completing the fault diagnosis of the current sensor by using the sensor fault estimation value and the characteristics thereof. According to the method, the sensor fault and the system state are accurately estimated, the switch tube and sensor faults can be jointly diagnosed, the fault diagnosis time is remarkably shortened, and the accuracy and robustness of switch tube fault diagnosis are improved.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fault diagnosis, and particularly relates to a combined fault diagnosis method for a three-phase four-bridge-arm wind power inverter. BACKGROUND

[0002] In wind power generation systems, three-phase four-bridge-arm inverters play an important role due to their unique structure and significant advantages. It has the characteristics of flexible topology structure, diverse control strategy, high output power quality and strong fault tolerance capability, which can effectively adapt to the complex power grid environment and load changes in wind power generation systems. By increasing the bridge arm, it can better suppress the zero sequence current, optimize the power quality, and at the same time has a certain redundancy design, which improves the reliability and fault tolerance capability of the system. However, due to the complex and changeable environment of wind power generation systems, the power devices and sensors in the inverter may still fail due to harsh environment or frequent load changes. Therefore, in order to ensure its reliability in actual operation, it is necessary to further improve the rapidity and accuracy of fault diagnosis to reduce the impact of faults on system operation, reduce maintenance costs, and ensure the stable operation of wind power generation systems.

[0003] The switch tube faults of three-phase four-bridge-arm inverters mainly include short circuit and open circuit. Short circuit fault usually damages the device due to the sharp increase of current, but can be disconnected in time by the protection circuit to prevent further damage; while open circuit fault may cause the control system to lose the correct current reference signal, and then cause output current distortion, and even cause system collapse in severe cases. In addition, current sensor faults, such as gain faults, sticking or open circuit, etc., will also cause the system to obtain incorrect current signals, affecting the output quality, and thus reducing the overall performance and reliability of the system. Therefore, fault diagnosis of the current sensor is also crucial, which can effectively ensure the stable operation of the inverter and ensure the safe and reliable operation of the system, prolong the service life of the equipment.

[0004] At present, most of the fault diagnosis technologies for inverters are aimed at switch tube faults, which can be roughly divided into the following types:

[0005] 1. Signal-based methods. These methods primarily process and analyze the raw electrical signals acquired in real time using specific analytical methods to extract information reflecting fault characteristics. Examples include the current space trajectory method and methods based on the Wave-U-Net neural network. Specific related papers and patents include "An On-Line Multiple Open-Circuit Fault Diagnostic Technique for Railway Vehicle Air-Conditioning Inverters" and "A Method and System for Diagnosing Open-Circuit Faults in Inverters" (Patent Application Publication No. CN116400264A). These methods require additional hardware, are computationally intensive, and require long diagnostic times.

[0006] 2. Artificial intelligence-based methods. These methods primarily simulate human thinking to achieve fault diagnosis. Examples include Fourier transform-based methods and convolutional neural network-based methods. Specific related papers and patents include "A Data-Driven Fault Diagnosis Methodology in Three-Phase Inverters for PMSM Drive Systems" and "A CNN-ISSA-SVM-Based Inverter Open-Circuit Fault Diagnosis Method" (Patent Application Publication No. CN119167188A). These methods require a certain number of real fault samples for the necessary machine learning and model training, resulting in high diagnostic complexity.

[0007] 3. Analytical model-based methods. These methods primarily establish a mathematical model of the inverter, compare the estimated system output with the measured information to obtain a residual, and analyze the residual signal to perform inverter fault diagnosis. Related papers and patents include "Inverter Open-Circuit Fault Diagnosis Method Based on Interval Sliding Mode Observer," "A Robust Observer-Based Method for IGBTs and Current Sensors Fault Diagnosis in Voltage-Source Inverters of PMSM Drives," and "A Model-Based IGBT Open-Circuit Fault Diagnosis Method for CRH3 EMU Traction Inverter" (Patent Application Publication No. CN105243258A). These methods rely on the accuracy of the established model.

[0008] In summary, existing technologies have problems such as difficulty in signal processing, large amount of calculation, high diagnostic complexity, and accuracy in model establishment. Summary of the Invention

[0009] The technical problem to be solved by the present invention is to address the above-mentioned drawbacks. The purpose of the present invention is to provide a joint fault diagnosis method for a three-phase, four-leg wind turbine inverter. By utilizing a non-singular matrix for decoupling, the method decomposes sensor faults and switch tube faults into two mutually non-interfering subsystems and establishes a reduced-order adaptive gain sliding mode observer to avoid the mutual influence between the two faults. By designing an adaptive detection threshold, the method reduces the fault detection time and improves the robustness of the fault diagnosis.

[0010] In order to achieve the above-mentioned object, the present invention provides a joint fault diagnosis method for a three-phase four-bridge-arm wind power inverter. The topology structure involved in the method includes a DC side power supply, a main inverter circuit, a three-phase LC filter, and a three-phase current sensor; the main inverter circuit includes four-phase bridge arms, wherein the fourth-phase bridge arm is denoted as the f-phase bridge arm, and the remaining three-phase bridge arms are denoted as the k-phase bridge arms, where k is the phase sequence of the three phases, k=a, b, c; in the four-phase bridge arms, each phase bridge arm is composed of two switching tubes connected in series, and the switching tube of the k-phase bridge arm is denoted as the switching tube V kj The switch tube of the f-phase bridge arm is the switch tube V fj , j=1,2; characterized in that the steps of the joint fault diagnosis method are as follows:

[0011] Step 1: sampling the three-phase output phase current and the three-phase output phase voltage to establish a hybrid logic dynamic model of the three-phase four-leg inverter;

[0012] Step 2: Establish the state equation of the three-phase four-leg inverter system containing switch tube faults and sensor faults, and then perform coordinate transformation, augmentation, and reconstruction to obtain subsystem 1 containing only switch tube open circuit faults and subsystem 2 containing only sensor faults;

[0013] Step 3: Design reduced-order adaptive gain sliding mode observers for subsystem 1 and subsystem 2 respectively, and obtain the k-phase current sensor fault estimation value through the two observers.

[0014] Step 4: Obtain the system switch fault residual e1 and the switch norm threshold Th0 based on the reduced-order adaptive gain sliding mode observer, and obtain the switch fault detection variable e0 according to the system switch fault residual e1;

[0015] Step 5: According to the fault diagnosis threshold ω of the k-phase current sensor k and k-phase current sensor fault estimation value Perform sensor fault diagnosis; according to the k-phase sensor fault detection characteristic quantity V k and k-phase fault location characteristic quantity W k To locate a sensor fault:

[0016] Step 6: Diagnose the switch tube fault based on the switch tube fault detection variable e0 and the switch tube fault detection threshold Th0, and locate the variable according to the upper half of the k-phase bridge arm switch tube fault. <T k >, lower half positioning variable <G k > and the k-phase bridge arm switch fault location threshold η k Locate the switch tube fault.

[0017] Preferably, the hybrid logic dynamic model of the three-phase four-leg inverter in step 1 is:

[0018]

[0019] Among them, i a ,i b ,i c is the three-phase output phase current, L is the inductance of the filter inductor in the three-phase LC filter, R is the equivalent resistance of the filter inductor in the three-phase LC filter, and u is the equivalent control input. Its expression is:

[0020]

[0021] Where u a ,u b ,u c is the three-phase output phase voltage, is the three-phase output phase voltage u a ,u b ,u c The first derivative of is the three-phase output phase voltage u a ,u b ,u c The second-order derivative of , C is the capacitance value of the filter capacitor in the three-phase LC filter, U dc is the voltage of the DC power supply, v f The dynamic switching function of the hybrid logic of the f-phase bridge arm is introduced, and the dynamic switching function of the hybrid logic of the k-phase bridge arm is introduced. k , k=a,b,c。

[0022] Preferably, the k-phase bridge arm hybrid logic dynamic switching function ν k and f-phase bridge arm mixed logic dynamic switching function v f Determine as follows:

[0023] The current flowing from the main inverter circuit to the load is defined as positive, and the current flowing from the load to the main inverter circuit is defined as negative. The logical variable μ of the k-phase bridge arm is defined as follows: k , μ k =1 means the k-phase current is positive, μ k =0 means the k-phase current is negative; define the logical variable μ of the f-phase bridge arm f, μ f =1 means the f phase current is positive, μ f =0 means the current of phase f is negative; the k-phase bridge arm switch V k1 and V k2 The switching signals are denoted as ξ k1 and ξ k2 , f-phase bridge arm switch tube V f1 and V f2 The switching signals are denoted as ξ f1 and ξ f2 , k-phase bridge arm hybrid logic dynamic switching function ν k and f-phase bridge arm mixed logic dynamic switching function v f The expressions are as follows:

[0024]

[0025]

[0026] The symbol “-” represents logical negation.

[0027] Preferably, the implementation process of step 2 is as follows:

[0028] Step 2.1: Establish the state equation of the three-phase four-leg inverter system with switch tube fault and sensor fault, and record it as its fault state equation, which is expressed as:

[0029]

[0030] Among them, y is the output signal of the system, x is the state variable of the system, is the derivative of the state variable x, A is the first coefficient matrix, B is the second coefficient matrix, B = diag[1 1 1] T , C is the third coefficient matrix, F a is the switch tube fault coefficient matrix, F s is the sensor fault coefficient matrix, D is the disturbance coefficient matrix, f a is the switch failure signal, f s is the sensor fault signal, d is the disturbance signal of the inverter;

[0031] Step 2.2: Use the first non-singular coordinate transformation matrix T and the second non-singular coordinate transformation matrix S to perform the following coordinate transformation on the fault state equation:

[0032]

[0033] Among them, A1, A2, A3, and A4 are the first decomposition submatrix, second decomposition submatrix, third decomposition submatrix, and fourth decomposition submatrix of the first coefficient matrix A, respectively. B1 and B2 are the first decomposition submatrix and second decomposition submatrix of the second coefficient matrix B, respectively. a1 is the switch tube fault coefficient matrix F a The decomposition submatrix of the perturbation coefficient matrix D is D1, D2 is the first decomposition submatrix and the second decomposition submatrix of the perturbation coefficient matrix D, C1, C4 is the first decomposition submatrix and the second decomposition submatrix of the third coefficient matrix C, F s2 is the sensor failure coefficient matrix F s The decomposition submatrix of ;

[0034] Introduce the coordinate transformation value of the state variable x And the coordinate transformation value of the output variable y The original fault state equation is decomposed into a primary subsystem 1 containing only the switch tube open circuit fault and a primary subsystem 2 containing only the sensor fault, where: x1 is the first decomposition state variable of state variable x, x2 is the second decomposition state variable of state variable x, y1 is the first decomposition signal of output signal y, and y2 is the second decomposition signal of output signal y; the expressions of primary subsystem 1 and primary subsystem 2 are:

[0035]

[0036] Where, is the derivative of the first decomposition state variable x1 of the state variable x, is the derivative of the second decomposition state variable x2 of the state variable x;

[0037] Step 2.3, augmentation and reconstruction

[0038] Subsystem 2 is obtained by augmenting the primary subsystem 2, and its expression is:

[0039]

[0040] in, is the augmentation of the second decomposition state variable x2 of the state variable x, for The derivative of is the augmented coefficient matrix 1, I2 is the unit matrix 2, is the augmented coefficient matrix 2, is the augmented coefficient matrix 3, I1 is the unit matrix 1, is the augmented coefficient matrix 4, is the augmented coefficient matrix 5, I3 is the unit matrix 3, is the augmented system disturbance coefficient matrix, is the augmented sensor fault coefficient matrix;

[0041] Reconstructing the primary subsystem 1 yields subsystem 1, which is expressed as:

[0042]

[0043] in, is the augmented coefficient matrix 6.

[0044] Preferably, the two observers in step 3 are: a reduced-order adaptive gain sliding mode observer designed for subsystem 1 and recorded as observer 1, and a reduced-order adaptive gain sliding mode observer designed for subsystem 2 and recorded as observer 2, and their expressions are as follows:

[0045]

[0046] in, is the estimated value of x1, for The derivative of for The estimated value of for The derivative of is the sensor fault signal f s The estimated value matrix of in,

[0047] is the estimated value of the a-phase current sensor fault, is the estimated value of the b-phase current sensor fault, is the estimated value of the c-phase current sensor fault, L1 is the feedback matrix of the subsystem 1 observer, L2 is the feedback matrix of the subsystem 2 observer, K1 is the gain of the subsystem 1 observer, K2 is the gain of the subsystem 2 observer, s1 is the sliding surface 1, s2 is sliding surface 2,

[0048] Introducing the k-phase current sensor fault estimation value k=a,b,c。

[0049] Preferably, the expression of the system switch failure residual e1 in step 4 is: Among them, e 1a is the residual error of phase a fault, e 1b is the residual error of phase b fault, e 1c is the c-phase fault residual; the switch tube fault detection variable e0 is introduced,

[0050] Preferably, the implementation process of step 5 is as follows:

[0051] Define the k-phase current sensor fault diagnosis threshold ω k ,ω k =i kmax ×5%,i kmax is the maximum value of the fundamental amplitude of the load current; define the k-phase sensor fault detection characteristic quantity V k and k-phase fault location characteristic quantity W k , in, is the k-phase output current;

[0052] The sensor fault diagnosis rules are:

[0053] like It is determined that the k-phase current sensor has no fault; if It is determined that the k-phase current sensor is faulty;

[0054] The rules for locating sensor faults are:

[0055] If V k =0 and W k =1, it is determined that the k-phase current sensor is stuck; if V k =0 and W k = -1, it is determined that the k-phase current sensor has an open circuit fault; when V k ≠0 and W k =1, it is determined that the k-phase current sensor has a gain fault.

[0056] Preferably, the implementation process of the switch tube open circuit fault diagnosis in step 6 is as follows:

[0057] Step 6.1, switch tube fault diagnosis, the rules are as follows:

[0058] If e0<Th0, it is determined that no switch tube open circuit fault has occurred, the system is normal, and the fault diagnosis is completed;

[0059] If e0 ≥ Th0, it is determined that the switch tube has an open circuit fault and the process goes to step 6.2;

[0060] Step 6.2, switch tube fault location

[0061] Step 6.21, first calculate the phase a fault residual e 1a , b-phase fault residual e 1b and c-phase fault residual e 1c Denoted as the k-phase bridge arm fault residual e 1k , k=a, b, c, then e 1k Divided into the upper half of the residual T k and the lower half residual G k, the expressions are:

[0062]

[0063] Then, the upper half of the positioning variable of the k-phase bridge arm switch tube fault is introduced <T k >Location variables for the lower half of the k-phase bridge arm switch failure <G k >, their expressions are:

[0064]

[0065] Wherein, T is the fundamental period of current, and t is the running time in the fundamental period of current T;

[0066] Finally, calculate the k-phase bridge arm switch fault location threshold η k , k=a, b, c; take η k 1% of the absolute value of the fundamental period of the current.

[0067] Step 6.2.2: Locate the switch fault according to the following rules:

[0068] like <T a ><η a , <T b >>η b , <T c >>η b , <G a >>η a , <gb><η b , <G c ><η c , then determine the switch tube V a1 An open circuit fault occurs;

[0069] like <T a >>η a , <T b ><η b , <T c ><η c , <G a ><η a , <G b >>η b , <G c >>η c , then determine the switch tube V a2 An open circuit fault occurs;

[0070] like <T a >>η a , <T b ><η b , <T c >>η c , <G a ><η a , <G b >>η b , <G c ><η c , then determine the switch tube V b1 An open circuit fault occurs;

[0071] like <T a ><η a , <T b >>η b , <T c ><η c , <G a >>η a , <G b ><η b , <G c >>η c Then determine the switch tube V b2 An open circuit fault occurs;

[0072] like <T a >>η a , <T b >>η b , <T c ><η c , <G a ><η a , <G b ><η b , <G c >>η c , then determine the switch tube V c1 An open circuit fault occurs;

[0073] like <T a ><η a , <T b ><η b , <T c >>η c , <G a >>η a , <G b >>η b , <G c ><η c , then determine the switch tube V c2 An open circuit fault occurs;

[0074] like <T a >>η a , <T b >>η b , <T c >>η c , <G a ><η a , <G b ><ηb , <G c ><η c , then determine the switch tube V f1 An open circuit fault occurs;

[0075] like <T a ><η a , <T b ><η b , <T c ><η c , <G a >>η a , <G b >>η b , <G c >>η c Then determine the switch tube V f2 An open circuit fault has occurred.

[0076] Preferably, the topological structure involved in the method further includes two DC capacitors, a control module and a three-phase load; the two DC capacitors are connected in series to form a DC capacitor branch and connected between the positive and negative busbars of the DC side power supply, and the main inverter circuit is connected in parallel with the DC capacitor branch; the three-phase LC filter includes a three-phase filter inductor and a three-phase filter capacitor, one end of the three-phase filter inductor is connected to the output end of the k-phase bridge arm, and the other end is connected to the three-phase load and the three-phase filter capacitor, and the other end of the three-phase load and the other end of the three-phase filter capacitor are connected to the output end of the k-phase bridge arm;

[0077] The detection end of the three-phase current sensor is connected between the three-phase filter inductor and the three-phase load, the output end is connected to the input of the control module, and the output end of the control module is respectively connected to the switch tubes of the four-phase bridge arms.

[0078] Compared with the prior art, the present invention has the following beneficial effects:

[0079] 1. The present invention uses non-singular coordinate transformation to decouple sensor faults from switch faults and establishes a reduced-order adaptive gain sliding mode observer, thereby avoiding the mutual influence of sensor faults and switch faults and achieving simultaneous diagnosis of inverter switch faults and sensor faults.

[0080] 2. The present invention selects an adaptive threshold for fault detection, which is resistant to disturbances and improves the accuracy and robustness of fault detection;

[0081] 3. The fault diagnosis process of the present invention does not require the addition of additional sensors, thereby reducing the cost of fault detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 is a topological diagram of a three-phase four-leg inverter according to an embodiment of the present invention;

[0083] Figure 2 is a flow chart of the combined fault diagnosis method of the present invention;

[0084] Figure 3 is the switch tube V in the embodiment of the present invention a1 Phase a output current i at 0.01 seconds a , a phase output current estimation value Switch tube fault detection variable e0 and switch tube fault detection threshold Th0, a phase current sensor fault estimation value and its a-phase current sensor fault diagnosis threshold ω a The simulation waveform diagram of

[0085] Figure 4 is the switch tube V in the embodiment of the present invention a1 Positioning variables of the switch failure when an open circuit failure occurs at 0.01 seconds <T k >、 <G k > and the k-phase bridge arm switch fault location threshold η k The simulation waveform diagram of

[0086] Figure 5 is the switch tube V in the embodiment of the present invention f1 Positioning variables of the switch failure when an open circuit failure occurs at 0.01 seconds <T k >、 <G k > and the k-phase bridge arm switch fault location threshold η k The simulation waveform diagram of

[0087] Figure 6 The a-phase current sensor S in the embodiment of the present invention is a When an open circuit fault occurs at 0.01 seconds, the output current i of phase A is a , a-phase current sensor fault estimation value and the fault diagnosis threshold ω of the phase a current sensor a , a-phase sensor fault classification characteristic value V a and W a The simulation waveform diagram of

[0088] Figure 7 In the embodiment of the present invention, when the a-phase current sensor is stuck at 0.0102, the a-phase output current i a , a-phase current sensor fault estimation value and the fault diagnosis threshold ω of the phase a current sensor a , a-phase sensor fault classification characteristic value V a and W a The simulation waveform diagram of

[0089] Figure 8 In the embodiment of the invention, when the phase a current sensor has a gain failure at 0.01 seconds, the output phase current i of phase a is a , a-phase current sensor fault estimation value and the fault diagnosis threshold ω of the phase a current sensor a , a-phase sensor fault classification characteristic value V a and W a The simulation waveform diagram of DETAILED DESCRIPTION

[0090] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0091] Figure 1 This is a circuit topology diagram of a three-phase four-bridge-arm wind power inverter in an example of the present invention. As can be seen from the figure, the topology structure involved in this method includes a DC side power supply, a main inverter circuit, a three-phase LC filter, and a three-phase current sensor; the main inverter circuit includes four-phase bridge arms, wherein the fourth-phase bridge arm is denoted as the f-phase bridge arm, and the remaining three-phase bridge arms are denoted as the k-phase bridge arms, where k is the phase sequence of the three phases, k = a, b, c; in the four-phase bridge arms, each phase bridge arm is composed of two switching tubes connected in series, and the switching tube of the k-phase bridge arm is denoted as the switching tube V kj The switch tube of the f-phase bridge arm is the switch tube V fj , j=1,2.

[0092] In this embodiment, the topology of the three-phase, four-arm wind turbine inverter involved in the method further includes two DC capacitors, a control module, and a three-phase load. The two DC capacitors are connected in series to form a DC capacitor branch, which is connected between the positive and negative busbars of the DC power supply. The main inverter circuit is connected in parallel with the DC capacitor branch. The three-phase LC filter includes a three-phase filter inductor and a three-phase filter capacitor. One end of the three-phase filter inductor is connected to the output end of the k-phase bridge arm, and the other end is connected to the three-phase load and the three-phase filter capacitor. The other end of the three-phase load and the other end of the three-phase filter capacitor are connected to the output end of the k-phase bridge arm.

[0093] The detection end of the three-phase current sensor is connected between the three-phase filter inductor and the three-phase load, the output end is connected to the input of the control module, and the output end of the control module is respectively connected to the switch tubes of the four-phase bridge arm.

[0094] C on the diagram 01 and C 02 is the DC capacitance, C 03 is a three-phase filter capacitor, D1-D8 are diodes, S a , S b , S c For phase a, phase b, and phase c current sensors, Z is a three-phase load. In this embodiment, U dc =360V, R=0.2Ω, L=150μH, C=48μF.

[0095] Figure 2 is a flow chart of the combined fault diagnosis method of the present invention, Figure 2 It can be seen that the combined fault diagnosis method includes sensor fault diagnosis and switch tube fault diagnosis, and specifically includes the following steps:

[0096] Step 1: Sample the three-phase output phase current and the three-phase output phase voltage to establish a hybrid logic dynamic model of the three-phase four-leg inverter.

[0097] In this embodiment, the hybrid logic dynamic model of the three-phase four-leg inverter in step 1 is:

[0098]

[0099] Among them, i a ,i b ,i c is the three-phase output phase current, L is the inductance of the filter inductor in the three-phase LC filter, R is the equivalent resistance of the filter inductor in the three-phase LC filter, and u is the equivalent control input. Its expression is:

[0100]

[0101] Where u a ,u b ,u c is the three-phase output phase voltage, is the three-phase output phase voltage u a ,u b ,u c The first derivative of is the three-phase output phase voltage u a ,u b ,u c The second-order derivative of , C is the capacitance value of the filter capacitor in the three-phase LC filter, U dc is the voltage of the DC power supply, v f The dynamic switching function of the hybrid logic of the f-phase bridge arm is introduced, and the dynamic switching function of the hybrid logic of the k-phase bridge arm is introduced. k , k=a,b,c。

[0102] In this embodiment, the k-phase bridge arm hybrid logic dynamic switching function ν k and f-phase bridge arm mixed logic dynamic switching function v f Determine as follows:

[0103] The current flowing from the main inverter circuit to the load is defined as positive, and the current flowing from the load to the main inverter circuit is defined as negative. The logical variable μ of the k-phase bridge arm is defined as follows: k , μ k =1 means the k-phase current is positive, μ k =0 means the k-phase current is negative; define the logical variable μ of the f-phase bridge arm f , μ f =1 means the f phase current is positive, μ f =0 means the current of phase f is negative; the k-phase bridge arm switch V k1 and V k2 The switching signals are denoted as ξ k1 and ξ k2 , f-phase bridge arm switch tube V f1 and V f2 The switching signals are denoted as ξ f1 and ξ f2 , k-phase bridge arm hybrid logic dynamic switching function ν k and f-phase bridge arm mixed logic dynamic switching function v f The expressions are as follows:

[0104]

[0105] The symbol "-" represents logical negation.

[0106] Step 2: Establish the state equation of the three-phase four-leg inverter system containing switch tube faults and sensor faults, and then perform coordinate transformation, augmentation and reconstruction to obtain subsystem 1 containing only switch tube open circuit faults and subsystem 2 containing only sensor faults.

[0107] In this embodiment, the implementation process of step 2 is as follows:

[0108] Step 2.1: Establish the state equation of the three-phase four-leg inverter system with switch tube fault and sensor fault, and record it as its fault state equation, which is expressed as:

[0109]

[0110] Among them, y is the output signal of the system, x is the state variable of the system, is the derivative of the state variable x, A is the first coefficient matrix, B is the second coefficient matrix, B = diag[1 1 1] T , C is the third coefficient matrix, C = diag[1 1 1] T , F a is the switch tube fault coefficient matrix, F s is the sensor fault coefficient matrix, D is the disturbance coefficient matrix, f a is the switch failure signal, f s is the sensor fault signal, d is the disturbance signal of the inverter;

[0111] Step 2.2: Use the first non-singular coordinate transformation matrix T and the second non-singular coordinate transformation matrix S to perform the following coordinate transformation on the fault state equation:

[0112]

[0113] Among them, A1, A2, A3, and A4 are the first decomposition submatrix, second decomposition submatrix, third decomposition submatrix, and fourth decomposition submatrix of the first coefficient matrix A, respectively. B1 and B2 are the first decomposition submatrix and second decomposition submatrix of the second coefficient matrix B, respectively. a1 is the switch tube fault coefficient matrix F a The decomposition submatrix of the perturbation coefficient matrix D is D1, B2 is the first decomposition submatrix and the second decomposition submatrix of the perturbation coefficient matrix D, C1, C4 is the first decomposition submatrix and the second decomposition submatrix of the third coefficient matrix C, F s2 is the sensor failure coefficient matrix F s The decomposition submatrix of ;

[0114] Introduce the coordinate transformation value of the state variable x And the coordinate transformation value of the output variable y The original fault state equation is decomposed into a primary subsystem 1 containing only the switch tube open circuit fault and a primary subsystem 2 containing only the sensor fault, where: x1 is the first decomposition state variable of state variable x, x2 is the second decomposition state variable of state variable x, y1 is the first decomposition signal of output signal y, and y2 is the second decomposition signal of output signal y; the expressions of primary subsystem 1 and primary subsystem 2 are:

[0115]

[0116] Where, is the derivative of the first decomposition state variable x1 of the state variable x, is the derivative of the second decomposition state variable x2 of the state variable x;

[0117] Step 2.3, augmentation and reconstruction

[0118] Subsystem 2 is obtained by augmenting the primary subsystem 2, and its expression is:

[0119]

[0120] in, is the augmentation of the second decomposition state variable x2 of the state variable x, for The derivative of is the augmented coefficient matrix 1, I2 is the unit matrix 2, is the augmented coefficient matrix 2, is the augmented coefficient matrix 3, I1 is the unit matrix 1, is the augmented coefficient matrix 4, is the augmented coefficient matrix 5, I3 is the unit matrix 3, is the augmented system disturbance coefficient matrix, is the augmented sensor fault coefficient matrix;

[0121] Reconstructing the primary subsystem 1 yields subsystem 1, which is expressed as:

[0122]

[0123] in, is the augmented coefficient matrix 6.

[0124] Step 3: Design reduced-order adaptive gain sliding mode observers for subsystem 1 and subsystem 2 respectively, and obtain the k-phase current sensor fault estimation value through the two observers.

[0125] In this embodiment, the two observers are: a reduced-order adaptive gain sliding mode observer designed for subsystem 1 and recorded as observer 1, and a reduced-order adaptive gain sliding mode observer designed for subsystem 2 and recorded as observer 2, and their expressions are as follows:

[0126]

[0127] in, is the estimated value of x1, for The derivative of for The estimated value of for The derivative of is the sensor fault signal f s The estimated value matrix of in,

[0128] is the estimated value of the a-phase current sensor fault, is the estimated value of the b-phase current sensor fault, is the estimated value of the c-phase current sensor fault, L1 is the feedback matrix of the subsystem 1 observer, L2 is the feedback matrix of the subsystem 2 observer, K1 is the gain of the subsystem 1 observer, K2 is the gain of the subsystem 2 observer, s1 is the sliding surface 1, s2 is sliding surface 2,

[0129] Introducing the k-phase current sensor fault estimation value k=a,b,c。

[0130] Step 4: Obtain the system switch fault residual e1 and the switch norm threshold Th0 based on the reduced-order adaptive gain sliding mode observer, and obtain the switch fault detection variable e0 according to the system switch fault residual e1.

[0131] In this embodiment, the expression of the system switch failure residual e1 is: Among them, e 1a is the residual error of phase a fault, e 1b is the residual error of phase b fault, e 1c is the c-phase fault residual; the switch tube fault detection variable e0 is introduced,

[0132] Step 5: Fault diagnosis of three-phase current sensor

[0133] According to the fault diagnosis threshold ω of the k-phase current sensor k and k-phase current sensor fault estimation value Perform sensor fault diagnosis; according to the k-phase sensor fault detection characteristic quantity V k and k-phase fault location characteristic quantity W k Perform sensor fault location.

[0134] In this embodiment, the implementation process of step 5 is as follows:

[0135] Define the k-phase current sensor fault diagnosis threshold ω k ,ω k =i kmax ×5%,i kmax is the maximum value of the fundamental amplitude of the load current; define the k-phase sensor fault detection characteristic quantity V k and k-phase fault location characteristic quantity W k , Among them, i k is the k-phase output current;

[0136] The sensor fault diagnosis rules are:

[0137] like It is determined that the k-phase current sensor has no fault; if It is determined that the k-phase current sensor is faulty;

[0138] The rules for locating sensor faults are:

[0139] If V k =0 and W k =1, it is determined that the k-phase current sensor is stuck; if V k =0 and W k = -1, it is determined that the k-phase current sensor has an open circuit fault; when V k ≠0 and W k =1, it is determined that the k-phase current sensor has a gain fault.

[0140] Step 6: Diagnose the open circuit fault of the switch tube

[0141] The switch tube fault diagnosis is performed based on the switch tube fault detection variable e0 and the switch tube fault detection threshold Th0, and the variable is located based on the upper half of the k-phase bridge arm switch tube fault. <T k >、Location variables of the lower half of the k-phase bridge arm switch failure <G k > and the k-phase bridge arm switch fault location threshold η k Locate the switch tube fault.

[0142] In this embodiment, the implementation process of the switch tube open circuit fault diagnosis in step 6 is as follows:

[0143] Step 6.1, switch tube fault diagnosis, the rules are as follows:

[0144] If e0<Th0, it is determined that no switch tube open circuit fault has occurred, the system is normal, and the fault diagnosis is completed;

[0145] If e0 ≥ Th0, it is determined that the switch tube has an open circuit fault and the process goes to step 6.2;

[0146] Step 6.2, switch tube fault location

[0147] Step 6.21, first calculate the phase a fault residual e 1a , b-phase fault residual e 1b and c-phase fault residual e 1c Denoted as the k-phase bridge arm fault residual e 1k , k=a, b, c, then e 1k Divided into the upper half of the residual T k and the lower half residual G k , the expressions are:

[0148]

[0149] Then, the upper half of the positioning variable of the k-phase bridge arm switch tube fault is introduced <T k >Location variables for the lower half of the k-phase bridge arm switch failure <G k >, their expressions are:

[0150]

[0151] Wherein, T is the fundamental period of current, and is the running time in the fundamental period of current T;

[0152] Finally, calculate the k-phase bridge arm switch fault location threshold η k , k=a, b, c; take η k 1% of the absolute value of the fundamental period of the current.

[0153] Step 6.2.2: Locate the switch fault according to the following rules:

[0154] like <T a ><η a , <T b >>η b , <T c >>η c , <G a >>η a , <G b ><η b , <G c ><η c , then determine the switch tube V a1 An open circuit fault occurs;

[0155] like <T a >>η a , <T b ><η b , <T c ><η c , <G a > <η a , <G b > > η b , <G c > < η c , then it is determined that the switch tube V a2 has an open circuit fault;

[0156] If <T a > > η a , <T b > < η b , <T c > > η c , <G a > < η a , <G b > > η b , <G c > < η c , then it is determined that the switch tube V b1 has an open circuit fault;

[0157] If <T a > < η a , <T b > > η b , <T c > < η c , <G a > > η a , <G b > < η b , <G c > > η c , then it is determined that the switch tube V b2 has an open circuit fault;

[0158] If <T a > > η a , <T b > > η b , <T c > < η c , <G a > < η a , <G b > < η b , <G c > > η c , then it is determined that the switch tube V c1 has an open circuit fault;

[0159] If <T a > < η a , <T b > < η b , <T c > > η c , <G a > > η a , <G b >>η b , <G c ><η c , then determine the switch tube V c2 An open circuit fault occurs;

[0160] like <T a >>η a , <T b >>η b , <T c >>η c , <G a ><η a , <G b ><η b , <G c ><η c , then determine the switch tube V f1 An open circuit fault occurs;

[0161] like <T a ><η a , <T b ><η b , <T c ><η c , <G a >>η a , <G b >>η b , <G c >>η c Then determine the switch tube V f2 An open circuit fault has occurred.

[0162] In order to demonstrate the beneficial effects of the present invention, a simulation of the present invention was performed. Figure 3 is the switch tube V in the embodiment of the present invention a1 Phase a output current i at 0.01 seconds a , a phase output current estimation value Switching tube fault detection variable e0 and its switching tube fault detection threshold Th0, a phase current sensor fault estimation value and its a-phase current sensor fault diagnosis threshold ω a The simulation waveform diagram of . It can be seen from the figure that after 0.01 seconds, i a There have been major changes, unchanged, e0 quickly exceeds Th0, No more than ω a .

[0163] Figure 4 is the switch tube V in the embodiment of the present invention a1 Positioning variables of the switch failure when an open circuit failure occurs at 0.01 seconds <T k >、 <G k > and the k-phase bridge arm switch fault location threshold η k The simulation waveform diagram. It can be seen from the figure that after 0.0105 seconds, <T a > does not exceed its threshold η a , <T b > exceeds its threshold η b , <T c > exceeds its threshold η b , <G a > exceeds its threshold η a , <G b > does not exceed its threshold η b , <G c > does not exceed its threshold η c , thus determining the switch tube V a1 An open circuit fault has occurred.

[0164] Figure 5 is the switch tube V in the embodiment of the present invention f1 Positioning variables of the switch failure when an open circuit failure occurs at 0.01 seconds <T k >、 <G k > and the k-phase bridge arm switch fault location threshold η k The simulation waveform diagram. It can be seen from the figure that after 0.0102 seconds, <T a > exceeds its threshold η a , <T b > exceeds its threshold η b , <T c > exceeds its threshold η c , <G a > does not exceed its threshold η a , <G b > does not exceed its threshold η b , <G c > does not exceed its threshold η c , thus determining the switch tube V f1 An open circuit fault has occurred.

[0165] Figure 6 The a-phase current sensor S in the embodiment of the present invention is a When an open circuit fault occurs at 0.01 seconds, the output current i of phase A is a , a-phase current sensor fault estimation value and its phase a current sensor fault diagnosis threshold ω a , a-phase sensor fault classification characteristic value V a , W a The simulation waveform diagram. It can be seen from the figure that after 0.0101 seconds, Rapidly exceeds its detection threshold ω a , V a =0 and W a =-1, thus determining that an open circuit fault has occurred in the a-phase sensor.

[0166] Figure 7 In the embodiment of the present invention, when the a-phase current sensor is stuck at 0.0102, the a-phase output current i a , a-phase current sensor fault estimation value and its a-phase current sensor fault diagnosis threshold ω a , a-phase sensor fault classification characteristic value V a , W a The simulation waveform diagram. It can be seen from the figure that after 0.0103, Rapidly surpass ω a , V a =0 and W a =1, thus determining that the phase a sensor has a stuck fault.

[0167] Figure 8 In the embodiment of the invention, when the phase a current sensor has a gain failure at 0.01 seconds, the output phase current i of phase a is a , a-phase current sensor fault estimation value and its a-phase current sensor fault diagnosis threshold ω a , a-phase sensor fault classification characteristic value V a , W a The simulation waveform diagram. As can be seen from the figure, at 0.0101 Rapidly surpass ω a , V a ≠0 and W a =1, thus determining that a phase a sensor has a gain failure.< / gb>

Claims

1. A three-phase four-bridge-arm wind turbine inverter joint fault diagnosis method, the method involving a topological structure including a DC side power supply, a main inverter circuit, a three-phase LC filter, and a three-phase current sensor; the main inverter circuit includes four-phase bridge arms, wherein the fourth-phase bridge arm is denoted as the f-phase bridge arm, and the remaining three-phase bridge arms are denoted as the k-phase bridge arms, where k is the phase sequence of the three phases, k = a, b, c; in the four-phase bridge arms, each phase bridge arm is composed of two switching tubes connected in series, and the switching tube of the k-phase bridge arm is denoted as the switching tube V kj The switch tube of the f-phase bridge arm is the switch tube V fj , j = 1, 2; characterized in that, The steps of the combined fault diagnosis method are as follows: Step 1: sampling the three-phase output phase current and the three-phase output phase voltage to establish a hybrid logic dynamic model of the three-phase four-leg inverter; Step 2: Establish the state equation of the three-phase four-leg inverter system containing switch tube faults and sensor faults, and then perform coordinate transformation, augmentation, and reconstruction to obtain subsystem 1 containing only switch tube open circuit faults and subsystem 2 containing only sensor faults; Step 3: Design reduced-order adaptive gain sliding mode observers for subsystem 1 and subsystem 2 respectively, and obtain the k-phase current sensor fault estimation value through the two observers. Step 4: Obtain the system switch fault residual e1 and the switch norm threshold Th0 based on the reduced-order adaptive gain sliding mode observer, and obtain the switch fault detection variable e0 according to the system switch fault residual e1; Step 5: According to the fault diagnosis threshold ω of the k-phase current sensor k and k-phase current sensor fault estimation value Perform sensor fault diagnosis; according to the k-phase sensor fault detection characteristic quantity V k and k-phase fault location characteristic quantity W k To locate a sensor fault: Step 6: Diagnose the switch tube fault based on the switch tube fault detection variable e0 and the switch tube fault detection threshold Th0, and locate the variable according to the upper half of the k-phase bridge arm switch tube fault. <T k >, lower half positioning variable <G k > and the k-phase bridge arm switch fault location threshold η k Locate the switch tube fault.

2. The three-phase four-bridge-arm wind power inverter joint fault diagnosis method according to claim 1, characterized in that: The hybrid logic dynamic model of the three-phase four-leg inverter in step 1 is: Among them, i a ,i b ,i c is the three-phase output phase current, L is the inductance of the filter inductor in the three-phase LC filter, R is the equivalent resistance of the filter inductor in the three-phase LC filter, and u is the equivalent control input. Its expression is: Where u a ,u b ,u c is the three-phase output phase voltage, is the three-phase output phase voltage u a ,u b ,u c The first derivative of is the three-phase output phase voltage u a ,u b ,u c The second-order derivative of , C is the capacitance value of the filter capacitor in the three-phase LC filter, U dc is the voltage of the DC power supply, v f The dynamic switching function of the hybrid logic of the f-phase bridge arm is introduced, and the dynamic switching function of the hybrid logic of the k-phase bridge arm is introduced. k , k=a,b,c。 3. A three-phase four-bridge-arm wind power inverter joint fault diagnosis method according to claim 2, characterized in that: The k-phase bridge arm hybrid logic dynamic switching function ν k and f-phase bridge arm mixed logic dynamic switching function v f Determine as follows: The current flowing from the main inverter circuit to the load is defined as positive, and the current flowing from the load to the main inverter circuit is defined as negative. The logical variable μ of the k-phase bridge arm is defined as follows: k , μ k =1 means the k-phase current is positive, μ k =0 means the k-phase current is negative; define the logical variable μ of the f-phase bridge arm f , μ f =1 means the f phase current is positive, μ f =0 means the current of phase f is negative; the k-phase bridge arm switch V k1 and V k2 The switching signals are denoted as ξ k1 and ξ k2 , f-phase bridge arm switch tube V f1 and V f2 The switching signals are denoted as ξ f1 and ξ f2 , k-phase bridge arm hybrid logic dynamic switching function ν k and f-phase bridge arm mixed logic dynamic switching function v f The expressions are as follows: The symbol "-" represents logical negation.

4. A three-phase four-bridge-arm wind power inverter joint fault diagnosis method according to claim 2, characterized in that: The implementation process of step 2 is as follows: Step 2.1: Establish the state equation of the three-phase four-leg inverter system with switch tube fault and sensor fault, and record it as its fault state equation, which is expressed as: Among them, y is the output signal of the system, x is the state variable of the system, is the derivative of the state variable x, A is the first coefficient matrix, B is the second coefficient matrix, B = diag[1 1 1] T , C is the third coefficient matrix, C = diag[1 1 1] T , F a is the switch tube fault coefficient matrix, F s is the sensor fault coefficient matrix, D is the disturbance coefficient matrix, f a is the switch failure signal, f s is the sensor fault signal, d is the disturbance signal of the inverter; Step 2.2: Use the first non-singular coordinate transformation matrix T and the second non-singular coordinate transformation matrix S to perform the following coordinate transformation on the fault state equation: Among them, A1, A2, A3, and A4 are the first decomposition submatrix, second decomposition submatrix, third decomposition submatrix, and fourth decomposition submatrix of the first coefficient matrix A, respectively. B1 and B2 are the first decomposition submatrix and second decomposition submatrix of the second coefficient matrix B, respectively. a1 is the switch tube fault coefficient matrix F a The decomposition submatrix of the perturbation coefficient matrix D is D1, D2 is the first decomposition submatrix and the second decomposition submatrix of the perturbation coefficient matrix D, C1, C4 is the first decomposition submatrix and the second decomposition submatrix of the third coefficient matrix C, F s 2 is the sensor failure coefficient matrix F s The decomposition submatrix of ; Introduce the coordinate transformation value of the state variable x And the coordinate transformation value of the output variable y The original fault state equation is decomposed into a primary subsystem 1 containing only the switch tube open circuit fault and a primary subsystem 2 containing only the sensor fault, where: x1 is the first decomposition state variable of state variable x, x2 is the second decomposition state variable of state variable x, y1 is the first decomposition signal of output signal y, and y2 is the second decomposition signal of output signal y; the expressions of primary subsystem 1 and primary subsystem 2 are: Where, is the derivative of the first decomposition state variable x1 of the state variable x, is the derivative of the second decomposition state variable x2 of the state variable x; Step 2.3, augmentation and reconstruction Subsystem 2 is obtained by augmenting the primary subsystem 2, and its expression is: in, is the augmentation of the second decomposition state variable x2 of the state variable x, for The derivative of is the augmented coefficient matrix 1, I2 is the unit matrix 2, is the augmented coefficient matrix 2, is the augmented coefficient matrix 3, I1 is the unit matrix 1, is the augmented coefficient matrix 4, is the augmented coefficient matrix 5, I3 is the unit matrix 3, is the augmented system disturbance coefficient matrix, is the augmented sensor fault coefficient matrix; Reconstructing the primary subsystem 1 yields subsystem 1, which is expressed as: in, is the augmented coefficient matrix 6.

5. A three-phase four-bridge-arm wind power inverter joint fault diagnosis method according to claim 4, characterized in that: The two observers in step 3 are: the reduced-order adaptive gain sliding mode observer designed for subsystem 1 and recorded as observer 1, and the reduced-order adaptive gain sliding mode observer designed for subsystem 2 and recorded as observer 2, and their expressions are as follows: in, is the estimated value of x1, for The derivative of for The estimated value of for The derivative of is the sensor fault signal f s The estimated value matrix of in, is the estimated value of the a-phase current sensor fault, is the estimated value of the b-phase current sensor fault, is the estimated value of the c-phase current sensor fault, L1 is the feedback matrix of the subsystem 1 observer, L2 is the feedback matrix of the subsystem 2 observer, K1 is the gain of the subsystem 1 observer, K2 is the gain of the subsystem 2 observer, s1 is the sliding surface 1, s2 is sliding surface 2, Introducing the k-phase current sensor fault estimation value k=a,b,c。 6. A three-phase four-bridge-arm wind power inverter joint fault diagnosis method according to claim 5, characterized in that: The expression of the system switch fault residual e1 in step 4 is: Among them, e 1a is the residual error of phase a fault, e 1b is the residual error of phase b fault, e 1c is the c-phase fault residual; the switch tube fault detection variable e0 is introduced, 7. A three-phase four-bridge-arm wind power inverter joint fault diagnosis method according to claim 6, characterized in that: The implementation process of step 5 is as follows: Define the k-phase current sensor fault diagnosis threshold ω k ,ω k =i kmax ×5%,i kmax is the maximum value of the fundamental amplitude of the load current; define the k-phase sensor fault detection characteristic quantity V k and k-phase fault location characteristic quantity W k , W k =sign(|i k |-ω k ), where i k is the k-phase output current; The sensor fault diagnosis rules are: like It is determined that the k-phase current sensor has no fault; if It is determined that the k-phase current sensor is faulty; The rules for locating sensor faults are: If V k =0 and W k =1, it is determined that the k-phase current sensor is stuck; if V k =0 and W k = -1, it is determined that the k-phase current sensor has an open circuit fault; when V k ≠0 and W k =1, it is determined that the k-phase current sensor has a gain fault.

8. A three-phase four-bridge-arm wind power inverter joint fault diagnosis method according to claim 7, characterized in that: The implementation process of the switch tube open circuit fault diagnosis in step 6 is as follows: Step 6.1, switch tube fault diagnosis, the rules are as follows: If e0<Th0, it is determined that no switch tube open circuit fault has occurred, the system is normal, and the fault diagnosis is completed; If e0 ≥ Th0, it is determined that the switch tube has an open circuit fault and the process goes to step 6.2; Step 6.2, switch tube fault location Step 6.21, first calculate the phase a fault residual e 1a , b-phase fault residual e 1b and c-phase fault residual e 1c Denoted as the k-phase bridge arm fault residual e 1k , k=a,b,c,then e 1k Divided into the upper half of the residual T k and the lower half residual G k , the expressions are: Then, the upper half of the positioning variable of the k-phase bridge arm switch tube fault is introduced <T k >Location variables for the lower half of the k-phase bridge arm switch failure <G k >, their expressions are: Wherein, T is the fundamental period of current, and t is the running time in the fundamental period of current T; Finally, calculate the k-phase bridge arm switch fault location threshold η k , k=a, b, c; take η k 1% of the absolute value of the fundamental period of the current. Step 6.2.2: Locate the switch fault according to the following rules: like <T a ><η a , <T b >>η b , <T c >>η c , <G a >>η a , <G b ><η b , <G c ><η c , then determine the switch tube V a1 An open circuit fault occurs; like <T a >>η a , <T b ><η b , <T c ><η c , <G a ><η a , <G b >>η b , <G c >>η c , then determine the switch tube V a2 An open circuit fault occurs; like <T a >>η a , <T b ><η b , <T c >>η c , <G a ><η a , <G b >>η b , <G c ><η c , then determine the switch tube V b1 An open circuit fault occurs; like <T a ><η a , <T b >>η b , <T c ><η c , <G a >>η a , <G b ><η b , <G c >>η c Then determine the switch tube V b2 An open circuit fault occurs; like <T a >>η a , <T b >>η b , <T c ><η c , <G a ><η a , <G b ><η b , <G c >>η c , then determine the switch tube V c1 An open circuit fault occurs; like <T a ><η a , <T b ><η b , <T c >>η c , <G a >>η a , <G b >>η b , <G c ><η c , then determine the switch tube V c2 An open circuit fault occurs; like <T a >>η a , <T b >>η b , <T c >>η c , <G a ><η a , <G b ><η b , <G c ><η c , then determine the switch tube V f1 An open circuit fault occurs; like <T a ><η a , <T b ><η b , <T c ><η c , <G a >>η a , <G b >>η b , <G c >>η c Then determine the switch tube V f2 An open circuit fault has occurred.

9. A three-phase four-bridge-arm wind power inverter joint fault diagnosis method according to claim 1, characterized in that: The topological structure involved in the method also includes two DC capacitors, a control module and a three-phase load; the two DC capacitors are connected in series to form a DC capacitor branch, and are connected between the positive and negative busbars of the DC side power supply; the main inverter circuit is connected in parallel with the DC capacitor branch; the three-phase LC filter includes a three-phase filter inductor and a three-phase filter capacitor, one end of the three-phase filter inductor is connected to the output end of the k-phase bridge arm, and the other end is connected to the three-phase load and the three-phase filter capacitor, and the other end of the three-phase load and the other end of the three-phase filter capacitor are connected to the output end of the k-phase bridge arm; The detection end of the three-phase current sensor is connected between the three-phase filter inductor and the three-phase load, the output end is connected to the input of the control module, and the output end of the control module is respectively connected to the switch tubes of the four-phase bridge arms.

Citation Information

Patent Citations

  • Model based IGBT open-circuit fault diagnosis method for traction inverter of CRH3 electric multiple unit

    CN105243258A

  • CNN-ISSA-SVM-based inverter open-circuit fault diagnosis method

    CN119167188A