A sensor fault diagnosis and fault-tolerant control method and system for a single-phase PWM rectifier considering non-ideal characteristics of switching devices

Abstract

This invention discloses a method and system for fault diagnosis and fault-tolerant control of a single-phase PWM rectifier sensor, considering the non-ideal characteristics of switching devices. The method includes: establishing a mathematical model of the single-phase PWM rectifier, transforming the non-ideal characteristics of the switching devices into a non-ideal switching function; establishing a state-space model of the single-phase PWM rectifier considering the non-ideal characteristics of the switching devices, designing an adaptive sliding mode observer, and tuning its parameters to provide an adaptive law for compensating for the ideal switching function; establishing fault residuals of the AC side current and DC side voltage of the single-phase PWM rectifier based on the adaptive sliding mode observer, performing fault diagnosis, locating the faulty sensor, and using the observation value of the adaptive sliding mode observer to replace the value collected by the faulty sensor for feedback and closed-loop control until the fault is eliminated, thus achieving fault-tolerant control. This invention improves the fault tolerance capability of the single-phase PWM rectifier under AC side current or DC side voltage sensor fault conditions.

Description

Technical Field

[0001] This invention relates to power electronic system control technology, and in particular to a single-phase PWM rectifier sensor fault diagnosis and fault-tolerant control method and system that takes into account the non-ideal characteristics of switching devices. Background Technology

[0002] Single-phase PWM rectifiers, employing an H-bridge topology consisting of four switching devices, can achieve AC-to-DC power conversion and have been widely used in high-speed railways, new energy vehicles, and related fields. However, in real-world industrial scenarios, the AC-side current sensors and DC-side voltage sensors used in single-phase PWM rectifiers can malfunction due to factors such as temperature rise, humidity changes, and static electricity. This can further cause the failure of the dual closed-loop control system for AC-side current and DC-side voltage in the single-phase PWM rectifier, jeopardizing the operational safety of the industrial system connected to the rectifier. Domestic and international researchers have conducted extensive research on sensor fault diagnosis and fault tolerance in power electronic systems. Existing methods are mainly divided into data-driven methods and analytical model-based methods.

[0003] The data-driven methods proposed in the literature (Q.Deng, B.Gou, X.Ge, et al. A High-Accuracy-Light-AI Data-Driven Diagnosis Method for Open-Circuit Faults in Single-Phase PWM Rectifiers[J].IEEE Transactions on Transportation Electrification,2023,9(3):4352-4365) can realize feature extraction and accurate detection of different fault modes in power electronic systems. Although the detection efficiency can be improved by reasonably designing the algorithm architecture, the implementation of fault-tolerant control still requires the help of other algorithms. Although the analytical model based on unknown input sliding mode observers proposed in the literature (J.Xia,Z.Li,X.Gao,et al.Real-Time Sensor Fault Identification and Remediation for Single-Phase Grid-Connected Converters Using Hybrid Observers With Unknown Input Adaptation[J].IEEE Transactions on Industrial Electronics,2023,70(3):2407-2418) is simple in structure, it depends on the accuracy of mathematical modeling. If the non-ideal characteristics of switching devices such as MOSFETs and IGBTs in actual power electronic systems are considered, the observation accuracy of state variables will be affected.

[0004] Analysis of current research both domestically and internationally reveals that data-driven methods cannot provide effective fault-tolerance solutions and are not easily implemented in single-phase PWM rectifier control in practical industrial scenarios. Methods based on analytical models, in establishing the switching functions of switching devices such as MOSFETs and IGBTs, do not consider non-ideal characteristics of switching devices such as on-resistance, on-state voltage drop, and turn-on / turn-off delay caused by junction temperature fluctuations, parasitic capacitance, voltage stress, and dead-time effects, thus their fault diagnosis accuracy needs improvement. Therefore, designing effective fault diagnosis and fault-tolerance schemes for single-phase PWM rectifier sensors, while considering the non-ideal characteristics of switching devices, has become a key factor in improving the survivability of single-phase PWM rectifiers in emergency situations. Summary of the Invention

[0005] Purpose of the invention: The purpose of this invention is to provide a method and system for fault diagnosis and fault-tolerant control of single-phase PWM rectifier sensors that takes into account the non-ideal characteristics of switching devices, so as to improve the fault tolerance capability of single-phase PWM rectifiers under AC side current or DC side voltage sensor fault conditions.

[0006] Technical solution: The fault diagnosis and fault-tolerant control method of the present invention includes the following steps:

[0007] A mathematical model of a single-phase PWM rectifier is established, and the non-ideal characteristics of the switching devices of the single-phase PWM rectifier are transformed into non-ideal switching functions based on the principle of area equivalence.

[0008] A state-space model of a single-phase PWM rectifier considering the non-ideal characteristics of switching devices is established, an adaptive sliding mode observer is designed, and based on the error dynamics of the adaptive sliding mode observer, the lower limit of the sign function amplitude is given. The Lyapunov function is selected to tune the parameters of the adaptive sliding mode observer, and an adaptive law for the compensation value of the ideal switching function is given.

[0009] Based on the fault residuals of the AC side current and DC side voltage of the single-phase PWM rectifier established by the adaptive sliding mode observer, fault diagnosis is performed. After locating the fault sensor, the AC side current or DC side voltage of the single-phase PWM rectifier observed online by the adaptive sliding mode observer is used to replace the value collected by the fault sensor for feedback and closed-loop control until the fault is eliminated, thus achieving fault-tolerant control.

[0010] Furthermore, the mathematical model of the single-phase PWM rectifier is established as follows:

[0011]

[0012] Where L is the AC side inductance; R is the AC side resistance; v g For AC side voltage; i g For AC side current; i L V is the load current; dc This is the DC side voltage; S A S B C represents the switching function of the two bridge arm switching devices; eq This is the equivalent capacitance on the DC side;

[0013] The switching function error caused by the non-ideal characteristics of the switching device is transformed into the turn-on / turn-off time. Then, using a method based on the area equivalence principle, the conduction time compensation term is further transformed into S. A In the function value, the switching function S after the second transformation cA Represented as:

[0014]

[0015] Among them, T s For the switching period, T onA Let m be the conduction time of switching device S1, where m is any positive integer, and δ is the conduction time of the switching device S1. A For the switching function S A The compensation value, where t is the sampling time;

[0016] Define the ideal switching function of the switching device as S0 = S A -S B Define the compensation value of the ideal switching function as δ = δ A -δ B , where δ B For the switching function S B The compensation value; through the above transformation method, the non-ideal characteristics of the switching device are ultimately determined by the non-ideal switching function S. c The depiction, specifically:

[0017]

[0018] Among them, S cB For S B The switching function after the second transformation, where k is any positive integer.

[0019] Furthermore, the state-space model of a single-phase PWM rectifier is represented as follows:

[0020]

[0021] in, Let be the time derivative of the state variable; x be the state variable; y be the output variable; u be the input variable; A be the state variable parameter matrix; B be the input parameter matrix; C be the output parameter matrix;

[0022] Design an adaptive sliding mode observer with the following state equation:

[0023]

[0024] in, For the observed values ​​of the state variables, For the time derivative of the observed values ​​of the state variable, Let F be the observed values ​​of the output variable, G be the gain matrix of the observation error of the output variable, and υ = ρsgn(e y ) represents the sign function term, ρ represents the magnitude of the sign function, and e y For the observation error of the output variable; A c This is a state variable parameter matrix that takes into account the non-ideal characteristics of the switching device.

[0025] Furthermore, based on the error dynamics of the adaptive sliding mode observer, a lower limit for the magnitude of the sign function is given; including:

[0026] The observer's error dynamic equation is:

[0027]

[0028] Where e is the observation error of the state variable, The time derivative of the observation error of the state variable. Let A be the observed values ​​of the state variables, F be the gain matrix of the state variable parameters, C be the output parameter matrix, G be the gain matrix of the sign function term, and υ = ρsgn(e y ) represents the sign function term, ρ represents the magnitude of the sign function, and e y The observation error of the output variable is ΔA = AA. c Let A be the error matrix of the state variable parameter matrix. c To consider the state variable parameter matrix of the switching device with respect to its non-ideal characteristics;

[0029] Decompose the observation error of the state variable into the observation error e of the observed variable. o The observation error e of the output variable y Define the coefficient matrix H = A - FC, H 11 H 12 H 21 H 22 These are the four elements of the coefficient matrix H; when the selected sign function magnitude satisfies ρ>|H 21 e o +H 22 e y |When, the observation error e of the output variable y Asymptotically converges to 0, |H 21 e o +H 22 e y This represents the lower limit of the magnitude of the sign function.

[0030] Furthermore, the Lyapunov function is selected to tune the parameters of the adaptive sliding mode observer, and an adaptive law for the compensation value of the ideal switching function is given; including:

[0031] The selected Lyapunov function V is:

[0032]

[0033] Among them, e T Let e ​​be the transpose of the state variable observation error, k1 and k2 be the positive coefficients of the Lyapunov function V, and A be the transpose of the state variable observation error. 12 With A 21Let A be the two elements on the second diagonal of matrix A, where A is the state variable parameter matrix.

[0034] The time derivative of the Lyapunov function V is expressed as:

[0035]

[0036] Where, δ v δ is the compensation value of the ideal switching function corresponding to the non-ideal switching function of the DC side voltage. i The compensation value for the ideal switching function corresponding to the non-ideal switching function of the AC side current is given, where L is the AC side inductance and C is the AC side inductance. eq Let F be the DC-side equivalent capacitance, F be the gain matrix of the output variable observation error, C be the output parameter matrix, and ΔA = AA. c Let A be the error matrix of the state variable parameter matrix. c To account for the non-ideal characteristics of the switching device, the state variable parameter matrix, These are the observed values ​​of the state variables;

[0037] To ensure that the time derivative formula of the Lyapunov function V is always less than 0, the adaptive law for the compensation values ​​of the ideal switching function corresponding to the AC side current and DC side voltage is obtained as follows:

[0038]

[0039] Among them, e y For the observation error of the output variable, e is the observed value of the alternating current. o The observation error of the observed quantity This is the observed value of the DC side voltage.

[0040] Furthermore, the fault types include: sensor value returning to zero, sudden gain change, sensor value drift, and sudden noise increase.

[0041] Furthermore, based on the fault residuals of the AC side current and DC side voltage of the single-phase PWM rectifier established by the adaptive sliding mode observer, fault diagnosis is performed. After locating the fault sensor, the AC side current or DC side voltage of the single-phase PWM rectifier observed online by the adaptive sliding mode observer is used to replace the fault sensor's collected values ​​for feedback and closed-loop control; including:

[0042] The observation errors of the AC side current and DC side voltage observations are normalized to obtain the fault residuals of the AC side current and DC side voltage.

[0043] In each sampling cycle of the single-phase PWM rectifier control, the fault residual is compared with the fault threshold. If the fault residual exceeds the fault threshold three times in five consecutive sampling cycles, the sensor of the physical quantity corresponding to the fault residual is considered to have failed.

[0044] After locating the faulty sensor, the sensor's collected values ​​are immediately isolated, and the observed values ​​obtained by the adaptive sliding mode observer are used for feedback and closed-loop control through the current controller, voltage controller, and sinusoidal pulse width modulation (SPWM) module.

[0045] The fault diagnosis and fault-tolerant control system corresponding to the method of the present invention includes:

[0046] The non-ideal characteristic conversion unit for switching devices is used to establish a mathematical model of a single-phase PWM rectifier and convert the non-ideal characteristics of the switching devices of the single-phase PWM rectifier into a non-ideal switching function based on the area equivalence principle.

[0047] An adaptive sliding mode observer construction unit is used to establish a state-space model of a single-phase PWM rectifier that considers the non-ideal characteristics of switching devices, design an adaptive sliding mode observer, give the lower limit of the sign function amplitude based on the error dynamics of the adaptive sliding mode observer, select the Lyapunov function to tune the parameters of the adaptive sliding mode observer, and give the adaptive law of the compensation value of the ideal switching function.

[0048] The sensor fault diagnosis and fault-tolerant control unit establishes the fault residuals of the AC side current and DC side voltage of the single-phase PWM rectifier based on the adaptive sliding mode observer, performs fault diagnosis, locates the faulty sensor, and uses the AC side current or DC side voltage of the single-phase PWM rectifier observed online by the adaptive sliding mode observer to replace the faulty sensor's collected value for feedback and closed-loop control until the fault is eliminated, thus achieving fault-tolerant control.

[0049] An electronic device for storing and executing the method, the device comprising:

[0050] Memory containing executable program code;

[0051] A processor coupled to the memory;

[0052] The processor calls the executable program code stored in the memory to execute the steps of the single-phase PWM rectifier sensor fault diagnosis and fault-tolerant control method that takes into account the non-ideal characteristics of switching devices.

[0053] A computer-readable storage medium for storing and executing the method, the computer-readable storage medium storing computer instructions, which, when invoked, are used to execute the steps of the single-phase PWM rectifier sensor fault diagnosis and fault-tolerant control method considering the non-ideal characteristics of switching devices.

[0054] Beneficial effects: Compared with the prior art, the significant technical effects of the present invention are: (1) There is no need to perform mechanism analysis and mathematical modeling on the high-order and nonlinear physical dynamics involved in the non-ideal characteristics of the switching device. The state variables under the condition of considering the non-ideal characteristics of the switching device can be accurately observed by using adaptive compensation, thereby reducing the complexity of the control algorithm; (2) The control architecture is simple, fault diagnosis and fault tolerance are convenient to implement, there is no need to collect a large amount of data in advance, and the complicated data processing links are eliminated, making it easy to deploy and execute in the digital signal processors commonly used in the industry. Attached Figure Description

[0055] Figure 1 This is a flowchart of the fault diagnosis and fault-tolerant control method described in this invention;

[0056] Figure 2 This invention relates to the single-phase PWM rectifier circuit structure.

[0057] Figure 3 This is a block diagram of the adaptive sliding mode observer structure of the present invention;

[0058] Figure 4 This is a block diagram of the fault diagnosis and fault-tolerant control system described in this invention;

[0059] Figure 5 The results are simulation results of the AC side current sensor gain change in a single-phase PWM rectifier. (a) shows the simulation results of the observed and acquired AC side current values, (b) shows the simulation results of the AC side current fault residual, (c) shows the simulation results of the fault sign, and (d) shows the simulation results of the actual DC side voltage value.

[0060] Figure 6 The results are simulation results of the DC-side voltage sensor in a single-phase PWM rectifier when the sensor value drifts. (a) shows the simulation results of the observed and acquired DC-side voltage values, (b) shows the simulation results of the DC-side voltage fault residual, (c) shows the simulation results of the fault sign, and (d) shows the simulation results of the actual AC-side current value. Detailed Implementation

[0061] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0062] like Figure 1As shown, the present invention provides a single-phase PWM rectifier sensor fault diagnosis and fault-tolerant control method considering the non-ideal characteristics of switching devices, comprising the following steps:

[0063] S1. Based on the mathematical model of a single-phase PWM rectifier, quantitatively characterize the non-ideal characteristics of the switching devices; including: establishing a mathematical model of the single-phase PWM rectifier, and transforming the non-ideal characteristics of the switching devices of the single-phase PWM rectifier into non-ideal switching functions based on the area equivalence principle; specifically:

[0064] The circuit structure of a single-phase PWM rectifier is as follows: Figure 2 As shown. Where L is the AC side inductance; R is the AC side resistance; L1 is the DC side filter inductance; C1 is the DC side filter capacitor; C0 is the DC side voltage regulator capacitor; v g For AC side voltage; i g For AC side current; i L V is the load current; dc S1 represents the DC side voltage; S2, S3, and S4 represent the switching states of the four switching devices; S A Let S be the switching function of the bridge arm containing switching devices S1 and S2. B This refers to the switching functions of the bridge arms containing switching devices S3 and S4. Figure 2 The mathematical model of a single-phase PWM rectifier can be obtained:

[0065]

[0066] Among them, C eq The DC-side equivalent capacitance, and the switching function S of the two bridge arm switching devices. A S B This can be specifically expressed as:

[0067]

[0068] Among them, T s For the switching period, T onA and T onB t represents the conduction time of switching devices S1 and S3, respectively, where m and n are arbitrary positive integers, and t is the sampling time.

[0069] As can be seen from equation (2), the switching function corresponding to the ideal characteristics of the switching device contains only two function values: 1 and 0.

[0070] To quantitatively characterize the non-ideal characteristics of switching devices, the switching function error caused by the non-ideal characteristics of the switching devices is first transformed into the turn-on / turn-off time, denoted as S. A For example, the switching function S′ after one transformation A It can be represented as:

[0071]

[0072] Among them, t eA This is the conduction time compensation item.

[0073] Due to the conduction time compensation term t eA Since direct measurement with sensors is not possible, a method based on the area equivalence principle is adopted to further transform the conduction time compensation term into S. A In the function value, the switching function S after the second transformation cA It can be represented as:

[0074]

[0075] Where, δ A For the switching function S A The compensation value.

[0076] Define the ideal switching function of the switching device as S0 = S A -S B Define the compensation value of the ideal switching function as δ = δ A -δ B , where δ B For the switching function S B The compensation value. Through the above transformation method, the non-ideal characteristics of the switching device can ultimately be derived from the non-ideal switching function S. c The depiction, specifically:

[0077]

[0078] Among them, S cB For S B After the second transformation, the switching function is δ, which is the compensation value of the ideal switching function, and k is any positive integer.

[0079] S2. Based on the state-space model of a single-phase PWM rectifier, construct an adaptive sliding mode observer; including: establishing a state-space model of the single-phase PWM rectifier considering the non-ideal characteristics of the switching devices, designing an adaptive sliding mode observer, selecting the Lyapunov function for parameter tuning, and deriving an adaptive law for the compensation value of the ideal switching function; specifically:

[0080] Define x as the state variable; y as the output variable; u as the input variable; matrix A as the state variable parameter matrix; matrix B as the input parameter matrix; and C as the output parameter matrix. Equation (1) can be expressed in the form of a state equation:

[0081]

[0082] in, Let be the time derivative of the state variable. Each vector and matrix in equation (6) can be specifically represented as:

[0083]

[0084] When observing different state variables, the value of matrix C in equation (6) is different, specifically:

[0085]

[0086] Based on equation (6), an adaptive sliding mode observer is constructed, and its state equation is:

[0087]

[0088] in, For the observed values ​​of the state variables, For the time derivative of the observed values ​​of the state variable, Let F be the observed values ​​of the output variable, G be the gain matrix of the observation error of the output variable, and υ = ρsgn(e y ) represents the sign function term, ρ represents the magnitude of the sign function, and e y For the observation error of the output variable, matrix A c The state variable parameter matrix considering the non-ideal characteristics of the switching device is specifically represented as follows:

[0089]

[0090] Among them, S cv and S ci Let be the non-ideal switching functions of the DC-side voltage and AC-side current, respectively, and let δ be the compensation values ​​of the corresponding ideal switching functions. v With δ i .

[0091] Taking the observation of the AC side current of a single-phase PWM rectifier as an example, the dynamic equation of the AC side current adaptive sliding mode observer is constructed as follows:

[0092]

[0093] in, The observed value is the alternating current. These are the observed values ​​of the DC-side voltage. Let be the time derivative of the observed value of the alternating current. The time derivative of the observed DC-side voltage is given by equation (11), where the parameter matrix A is... c The elements of B are respectively A c11 =-R / L, A c12 =-S cv / L, A c21 =S ci / C eqA c22 =0, B 11 =1 / L, B 22 =-1 / C eq The coefficients g1, g2 and f1, f2 are elements of matrices G and F, respectively.

[0094] Based on equation (11), the online observation of the AC side current of the single-phase PWM rectifier can be realized. The dynamic equation structure of the DC side voltage adaptive sliding mode observer of the single-phase PWM rectifier is the same, except that the structure of matrix C is different. The selection of matrix C requires the use of equation (8).

[0095] Parameter tuning for the adaptive sliding mode observer includes: providing a lower limit for the sign function amplitude based on the error dynamics of the adaptive sliding mode observer, establishing a Lyapunov function, and providing an adaptive law for the compensation value of the ideal switching function.

[0096] Subtracting equation (6) from equation (9), we obtain the following error dynamic equation:

[0097]

[0098] Where e is the observation error of the state variable, Let ΔA be the time derivative of the observation error of the state variable. c The error matrix is ​​the parameter matrix of the state variables.

[0099] Decompose the observation error of the state variable into the observation error e of the observed variable. o The observation error e of the output variable y Define a 2×2 coefficient matrix H = A - FC, H 11 H 12 H 21 H 22 Let H be the four elements of a 2×2 matrix. The decomposed error dynamic equation is:

[0100]

[0101] in, The time derivative of the observation error of the observed quantity. The time derivative of the observation error of the output variable, with coefficient g. o =g1.

[0102] According to equation (13), the following inequality can be obtained.

[0103]

[0104] When the magnitude of the selected symbol function satisfies ρ>|H 21 e o +H22 e y |When, the observation error e of the output variable y It will asymptotically converge to 0, |H 21 e o +H 22 e y This represents the lower limit of the magnitude of the sign function.

[0105] When the observation error e of the output variable y After converging to 0, equation (12) becomes Choose the following Lyapunov function V:

[0106]

[0107] Where k1 and k2 are the positive coefficients of the Lyapunov function V, A 12 With A 21 The two elements on the second diagonal of matrix A are -S0 / L and S0 / C. eq .

[0108] The time derivative of the Lyapunov function V can be expressed as:

[0109]

[0110] Among them, e T δ is the transpose of the observation error of the state variable. v δ is the compensation value of the ideal switching function corresponding to the non-ideal switching function of the DC side voltage. i This is the compensation value for the ideal switching function corresponding to the non-ideal switching function of the AC side current. It can be obtained using the following formula:

[0111]

[0112] When the pole placement method is used to select matrix F, we have e T [(A-FC) T +(A-FC)]e<0 holds true. To ensure that equation (16) is always less than 0, according to equation (17), we can obtain:

[0113]

[0114] Simplifying, we obtain the adaptive law for the compensation values ​​of the ideal switching functions corresponding to the AC side current and DC side voltage:

[0115]

[0116] The structural block diagram of the adaptive sliding mode observer of this invention is as follows: Figure 3As shown, equation (19) is replaced by a proportional-integral (PI) regulator.

[0117] S3. An adaptive sliding mode observer is used to observe the AC side current and DC side voltage of the single-phase PWM rectifier online. Fault diagnosis is performed based on the fault residuals and fault thresholds of the AC side current and DC side voltage. After locating the fault sensor, the sensor's data is isolated. The observed values ​​obtained by the adaptive sliding mode observer are then used for feedback and closed-loop control through a current controller, voltage controller, and sinusoidal pulse width modulation (SPWM) module until the fault is cleared, thus achieving fault-tolerant control. Specifically:

[0118] This invention considers four types of faults in the AC side current and DC side voltage sensors of a single-phase PWM rectifier: zero sensing value, sudden gain change, sensing value drift, and sudden noise increase.

[0119] By normalizing the observation errors of the AC side current and DC side voltage observations obtained from equation (11), the fault residuals of the AC side current and DC side voltage can be obtained.

[0120] In each sampling cycle of the single-phase PWM rectifier control, the fault residual is compared with the fault threshold. If the fault residual exceeds the fault threshold three times in five consecutive sampling cycles, the sensor of the physical quantity corresponding to the fault residual is considered to have failed.

[0121] After locating the faulty sensor, the sensor's collected values ​​are immediately isolated, and closed-loop control is performed using the observation values ​​obtained from the adaptive sliding mode observer until the fault is eliminated. Fault tolerance can be achieved in this way.

[0122] like Figure 4 As shown, the fault diagnosis and fault-tolerant control system used in the above control method includes:

[0123] The non-ideal characteristic conversion unit for switching devices is used to quantitatively characterize the non-ideal characteristics of switching devices based on the mathematical model of a single-phase PWM rectifier. Specifically, it establishes a mathematical model of a single-phase PWM rectifier and converts the non-ideal characteristics of the switching devices of the single-phase PWM rectifier into non-ideal switching functions based on the principle of area equivalence.

[0124] An adaptive sliding mode observer construction unit is used to construct an adaptive sliding mode observer based on the state-space model of a single-phase PWM rectifier. Specifically, it establishes a state-space model of a single-phase PWM rectifier considering the non-ideal characteristics of the switching devices based on the non-ideal switching function, designs an adaptive sliding mode observer, gives the lower limit of the sign function amplitude based on the error dynamics of the adaptive sliding mode observer, establishes the Lyapunov function, and gives the adaptive law for the compensation value of the ideal switching function; and observes the AC side current and DC side voltage of the single-phase PWM rectifier online.

[0125] The sensor fault diagnosis and fault-tolerant control unit includes a sensor fault diagnosis and fault-tolerant control module, an AC side current controller module, and a sinusoidal pulse width modulation (SPWM) module. The sensor fault diagnosis and fault-tolerant control module is used to observe the AC side current and DC side voltage of the single-phase PWM rectifier online using an adaptive sliding mode observer. Fault diagnosis is performed based on the fault residual and fault threshold of the AC side current and DC side voltage. After locating the faulty sensor, the sensor's acquired value is isolated, and the observed value obtained by the adaptive sliding mode observer is used to perform feedback and closed-loop control through the AC side current controller module, AC side current controller module, and sinusoidal pulse width modulation (SPWM) module until the fault is eliminated, thus achieving fault-tolerant control.

[0126] When both the AC-side current sensor and the DC-side voltage sensor of the single-phase PWM rectifier are working normally, the closed-loop control of the single-phase PWM rectifier can be directly performed using the measured values ​​from the AC-side current sensor and the DC-side voltage sensor. During normal operation, the system mainly consists of an AC-side current controller module, a DC-side voltage controller module, and an SPWM module to realize the AC-to-DC power conversion function of the single-phase PWM rectifier.

[0127] When the AC-side current sensor or DC-side voltage sensor of a single-phase PWM rectifier fails, the adaptive sliding mode observer can adaptively compensate for modeling errors while considering the non-ideal characteristics of the switching devices, achieving accurate observation of the AC-side current or DC-side voltage. The AC-side current or DC-side voltage observation values ​​obtained from the adaptive sliding mode observer module are fed into the sensor fault diagnosis and fault-tolerant control module. The observation error is compared with a threshold to locate the fault, and the faulty sensor's collected value is replaced for feedback and closed-loop control. The resulting control signals, i.e., the switching signals of the four switching devices, are provided by the SPWM module to realize the AC-to-DC power conversion function of the single-phase PWM rectifier under fault conditions.

[0128] The method proposed in this invention is robust and can adaptively compensate for modeling errors while taking into account the non-ideal characteristics of switching devices. It enables accurate observation of AC side current and DC side voltage of single-phase PWM rectifier, as well as accurate diagnosis and effective fault tolerance of four types of sensor faults.

[0129] An electronic device according to the present invention, the device comprising:

[0130] Memory containing executable program code;

[0131] A processor coupled to the memory;

[0132] The processor calls the executable program code stored in the memory to execute the steps of the single-phase PWM rectifier sensor fault diagnosis and fault-tolerant control method that takes into account the non-ideal characteristics of switching devices, as described above.

[0133] The present invention provides a computer-readable storage medium storing computer instructions, which, when invoked, are used to execute the steps of the single-phase PWM rectifier sensor fault diagnosis and fault-tolerant control method as described above, taking into account the non-ideal characteristics of switching devices.

[0134] The effectiveness of the proposed single-phase PWM rectifier sensor fault diagnosis and fault-tolerant control method and system, which considers the non-ideal characteristics of switching devices, is verified through MATLAB / Simulink simulation experiments. The simulation results are presented below. The single-phase PWM rectifier parameters selected in the simulation experiments are as follows:

[0135] The AC side inductance L is 5mH, the AC side resistance R is 0.2Ω, the DC side voltage regulator C0 is 2200μF, the load resistance is 20~25Ω, the DC side filter inductance L1 is 1.15mH, the DC side filter capacitor C1 is 2200μF, and the AC side voltage v g The effective value is 220V, and the DC side voltage is v. dc The given value is 400V; the IGBT modulation frequency is 10kHz, the dead time is 4μs, the IGBT on-resistance is 0.1Ω, the IGBT on-state voltage drop is 2.1V, and the diode on-state voltage drop is 0.4V.

[0136] In the first set of simulation experiments, a sudden load change was simulated at t = 0.3s, from 20Ω to 25Ω. A sudden gain change in the AC side current sensor was simulated at t = 0.407s, with the gain of the fault sensor's measured value changed to 0.2. The simulation results are as follows: Figure 5 As shown in (a), (b), (c), and (d) in the second set of simulation experiments, a sudden load change was simulated at t = 0.3s, from 20Ω to 25Ω, and a drift in the DC-side voltage sensor value was simulated at t = 0.407s, by reducing the faulty sensor's measured value by 320V. The simulation results are as follows. Figure 6As shown in (a), (b), (c), and (d), the simulation results demonstrate that the single-phase PWM rectifier sensor fault diagnosis and fault tolerance method proposed in this invention can detect sensor faults within five sampling periods after a fault occurs. After replacing the faulty sensor's collected values ​​with the observation values ​​obtained from the adaptive sliding mode observer, the single-phase PWM rectifier maintains normal operation and is not affected by the sensor fault. The simulation results confirm that the adaptive sliding mode observer proposed in this invention can accurately observe the AC side current and DC side voltage of the single-phase PWM rectifier under the condition of considering the non-ideal characteristics of the switching devices. The proposed sensor fault diagnosis and fault tolerance control method can accurately locate sensor faults and quickly implement fault tolerance, thereby improving the reliability of the single-phase PWM rectifier operation.

Claims

1. A method for sensor fault diagnosis and fault-tolerant control of a single-phase PWM rectifier considering the non-ideal characteristics of switching devices, characterized in that, Includes the following steps: A mathematical model of a single-phase PWM rectifier is established. Based on the principle of area equivalence, the non-ideal characteristics of the switching devices in the single-phase PWM rectifier are transformed into non-ideal switching functions. The established mathematical model of the single-phase PWM rectifier is as follows: ； in, For AC side inductance; For AC side resistance; This is the AC side voltage; This is the alternating current. This is the load current; This is the DC side voltage; , For the switching functions of the two bridge arm switching devices; This is the equivalent capacitance on the DC side; The switching function error caused by the non-ideal characteristics of the switching device is transformed into the turn-on / turn-off time. Then, a method based on the area equivalence principle is used to further transform the conduction time compensation term into... In the function value, the switching function after the second transformation Represented as: ； in, For the switching cycle, Switching devices On-time, For any positive integer, For switching functions The compensation value, Sampling time; Define the ideal switching function of a switching device Define the compensation value of the ideal switching function. ,in, For switching functions The compensation value; through the above transformation method, the non-ideal characteristics of the switching device are ultimately determined by the non-ideal switching function. The depiction, specifically: ； in, for The switching function after secondary transformation It can be any positive integer; A state-space model of a single-phase PWM rectifier considering the non-ideal characteristics of switching devices is established, an adaptive sliding mode observer is designed, and based on the error dynamics of the adaptive sliding mode observer, the lower limit of the sign function amplitude is given. The Lyapunov function is selected to tune the parameters of the adaptive sliding mode observer, and an adaptive law for the compensation value of the ideal switching function is given. Based on the fault residuals of the AC side current and DC side voltage of the single-phase PWM rectifier established by the adaptive sliding mode observer, fault diagnosis is performed. After locating the fault sensor, the AC side current or DC side voltage of the single-phase PWM rectifier observed online by the adaptive sliding mode observer is used to replace the value collected by the fault sensor for feedback and closed-loop control until the fault is eliminated, thus achieving fault-tolerant control.

2. The single-phase PWM rectifier sensor fault diagnosis and fault-tolerant control method considering the non-ideal characteristics of switching devices according to claim 1, characterized in that, The state-space model of a single-phase PWM rectifier is represented as follows: ； in, The time derivative of the state variable; For state variables; For output variables; For input variables; This is the state variable parameter matrix; The input parameter matrix; This is the output parameter matrix; Design an adaptive sliding mode observer with the following state equation: ； in, For the observed values ​​of the state variables, For the time derivative of the observed values ​​of the state variable, For the observed values ​​of the output variables, The gain matrix is ​​the observation error of the output variable. This is the gain matrix of the sign function term; For symbolic function terms, The sign function magnitude. The observation error of the output variable; This is a state variable parameter matrix that takes into account the non-ideal characteristics of the switching device.

3. The single-phase PWM rectifier sensor fault diagnosis and fault-tolerant control method considering the non-ideal characteristics of switching devices according to claim 1, characterized in that, Based on the error dynamics of the adaptive sliding mode observer, the lower bound of the sign function amplitude is given; including: The observer's error dynamic equation is: ； Where e is the observation error of the state variable, The time derivative of the observation error of the state variable. Let A be the observed values ​​of the state variables, A be the parameter matrix of the state variables, and F be the gain matrix of the observation errors of the output variables. G is the output parameter matrix, and G is the gain matrix of the sign function term. For symbolic function terms, The sign function magnitude. For the observation error of the output variable, The error matrix is ​​the parameter matrix of the state variables. To consider the state variable parameter matrix of the switching device with respect to its non-ideal characteristics; Decompose the observation error of the state variable into the observation error of the observed variable. Observation error of output variable Define the coefficient matrix H = A - FC, H 11 H 12 H 21 H 22 These are the four elements of the coefficient matrix H; when the selected sign function magnitude satisfies At that time, the observation error of the output variable Asymptotically converges to 0. This is the lower limit of the magnitude of the sign function.

4. The single-phase PWM rectifier sensor fault diagnosis and fault-tolerant control method considering the non-ideal characteristics of switching devices according to claim 1, characterized in that, The Lyapunov function is selected for parameter tuning of the adaptive sliding mode observer, and the adaptive law for the compensation value of the ideal switching function is given; including: Selected Lyapunov function for: ； in, Let e ​​be the transpose of the state variable observation error. and For Lyapunov functions The positive coefficient, and For matrix The two elements on the second diagonal, A, represent the state variable parameter matrix; Lyapunov function The time derivative is expressed as: ； in, This is the compensation value for the ideal switching function corresponding to the non-ideal switching function of the DC-side voltage. This is the compensation value for the ideal switching function corresponding to the non-ideal switching function of the AC side current. For AC side inductance, Let F be the equivalent capacitance on the DC side, and F be the gain matrix of the observation error of the output variable. For the output parameter matrix, The error matrix is ​​the parameter matrix of the state variables. To account for the non-ideal characteristics of the switching device, the state variable parameter matrix, These are the observed values ​​of the state variables; To make Lyapunov functions Since the time derivative formula is always less than 0, the adaptive law for the compensation values ​​of the ideal switching function corresponding to the AC side current and DC side voltage is: ； in, For the observation error of the output variable, The observed value is the alternating current. The observation error of the observed quantity This is the observed value of the DC side voltage.

5. The single-phase PWM rectifier sensor fault diagnosis and fault-tolerant control method considering the non-ideal characteristics of switching devices according to claim 1, characterized in that, Fault types include: sensor value returning to zero, sudden gain change, sensor value drift, and sudden noise increase.

6. The single-phase PWM rectifier sensor fault diagnosis and fault-tolerant control method considering the non-ideal characteristics of switching devices according to claim 1, characterized in that, Based on the fault residuals of the AC side current and DC side voltage of the single-phase PWM rectifier established by the adaptive sliding mode observer, fault diagnosis is performed. After locating the fault sensor, the AC side current or DC side voltage of the single-phase PWM rectifier observed online by the adaptive sliding mode observer is used to replace the fault sensor's acquired values ​​for feedback and closed-loop control; including: The observation errors of the AC side current and DC side voltage observations are normalized to obtain the fault residuals of the AC side current and DC side voltage. In each sampling cycle of the single-phase PWM rectifier control, the fault residual is compared with the fault threshold. If the fault residual exceeds the fault threshold three times in five consecutive sampling cycles, the sensor of the physical quantity corresponding to the fault residual is considered to have failed. After locating the faulty sensor, the sensor's collected values ​​are immediately isolated, and the observed values ​​obtained by the adaptive sliding mode observer are used for feedback and closed-loop control through the current controller, voltage controller, and sinusoidal pulse width modulation module.

7. A single-phase PWM rectifier sensor fault diagnosis and fault-tolerant control system considering the non-ideal characteristics of switching devices, characterized in that, include: The non-ideal characteristic conversion unit for switching devices is used to establish the mathematical model of a single-phase PWM rectifier. Based on the area equivalence principle, it converts the non-ideal characteristics of the switching devices in the single-phase PWM rectifier into a non-ideal switching function. The established mathematical model of the single-phase PWM rectifier is as follows: ； in, For AC side inductance; For AC side resistance; This is the AC side voltage; This is the alternating current. This is the load current; This is the DC side voltage; , For the switching functions of the two bridge arm switching devices; This is the equivalent capacitance on the DC side; The switching function error caused by the non-ideal characteristics of the switching device is transformed into the turn-on / turn-off time. Then, a method based on the area equivalence principle is used to further transform the conduction time compensation term into... In the function value, the switching function after the second transformation Represented as: ； in, For the switching cycle, Switching devices On-time, For any positive integer, For switching functions The compensation value, Sampling time; Define the ideal switching function of a switching device Define the compensation value of the ideal switching function. ,in, For switching functions The compensation value; through the above transformation method, the non-ideal characteristics of the switching device are ultimately determined by the non-ideal switching function. The depiction, specifically: ； in, for The switching function after secondary transformation It can be any positive integer; An adaptive sliding mode observer construction unit is used to establish a state-space model of a single-phase PWM rectifier that considers the non-ideal characteristics of switching devices, design an adaptive sliding mode observer, give the lower limit of the sign function amplitude based on the error dynamics of the adaptive sliding mode observer, select the Lyapunov function to tune the parameters of the adaptive sliding mode observer, and give the adaptive law of the compensation value of the ideal switching function. The sensor fault diagnosis and fault-tolerant control unit establishes the fault residuals of the AC side current and DC side voltage of the single-phase PWM rectifier based on the adaptive sliding mode observer, performs fault diagnosis, locates the faulty sensor, and uses the AC side current or DC side voltage of the single-phase PWM rectifier observed online by the adaptive sliding mode observer to replace the faulty sensor's collected value for feedback and closed-loop control until the fault is eliminated, thus achieving fault-tolerant control.

8. An electronic device, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the steps of the single-phase PWM rectifier sensor fault diagnosis and fault-tolerant control method considering the non-ideal characteristics of switching devices as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, which, when invoked, are used to execute the steps of the single-phase PWM rectifier sensor fault diagnosis and fault-tolerant control method considering the non-ideal characteristics of switching devices as described in any one of claims 1-6.

Images