A sensor fault handling method, apparatus, and medium

Abstract

This application discloses a sensor fault handling method, apparatus, and medium, relating to the field of power supply technology for auxiliary converters in rail transit. State estimation equations are constructed based on each sampled data and the corresponding sensor's position in the auxiliary converter to obtain state estimates. The coupling relationships between sensors in the auxiliary converter are established based on these state estimation equations to determine the current reference sensor. The coupling relationships between sensors at different positions in the multi-stage converter within the auxiliary converter are then combined to decouple and determine the state of the reference sensor, facilitating subsequent forward and backward sequential diagnosis. The fault handling results for each sensor are determined based on the sampled data of the current reference sensor and the state estimates of its adjacent sensors, as well as the sampled data of the other sensors (excluding the current reference sensor) and their corresponding state estimates, achieving precise multi-sensor positioning.

Description

Technical Field

[0001] This application relates to the field of power supply technology for auxiliary converters in rail transit, and in particular to a sensor fault handling method, device and medium. Background Technology

[0002] As the interface between the power grid and on-board equipment, the converter relies heavily on sensors for its normal and reliable operation. A typical auxiliary converter system contains several or even dozens of current and voltage sensors. Auxiliary converters often have a multi-stage structure and a large number of sensors with interconnected data acquisition states. Currently, common sensor diagnostics rely on the method of cross-verification between adjacent sensors. This involves calculating the estimated value of sensor B based on the sampled value of sensor A and comparing it with the sampled value of sensor B. The deviation level is then used to determine whether sensor B has failed. However, this state estimation is reciprocal, simultaneously using the sampled value of sensor B to diagnose sensor A. Therefore, when sensor B fails, both sensor A and B may be reported as faulty simultaneously, making it difficult to pinpoint the specific sensor failure and reducing the reliability of the auxiliary converter.

[0003] Therefore, how to accurately locate sensor faults in auxiliary converters is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] The purpose of this application is to provide a sensor fault handling method, device, and medium to solve the problem of positioning difficulties and reduced reliability of auxiliary converters caused by the use of dual sensor mutual detection in conventional solutions.

[0005] To address the aforementioned technical problems, this application provides a sensor fault handling method, comprising:

[0006] Acquire sampling data from each sensor of the auxiliary converter;

[0007] Based on the sampling data and the corresponding sensor positions in the auxiliary converter, a state estimation equation is constructed to obtain the state estimate value;

[0008] The coupling relationship between the sensors in the auxiliary converter is established based on the state estimation equations to determine the current reference sensor.

[0009] In the coupling relationship, the fault handling result of each sensor is determined based on the sampling data of the current reference sensor and the state estimate value of each adjacent sensor, and the sampling data of the other sensors besides the current reference sensor and the state estimate value of each corresponding adjacent sensor, so as to complete the fault location processing.

[0010] On one hand, the sampled data includes at least input voltage, input current, boost voltage, intermediate voltage, first bridge arm current, second bridge arm current, first output voltage, and second output voltage; based on each of the sampled data and the corresponding sensor positions on the auxiliary converter, a state estimation equation is constructed to obtain the state estimate, including:

[0011] When the boost converter of the auxiliary converter enters the continuous state, the state estimation equation is constructed based on the duty cycle of the switching devices in the auxiliary converter to obtain the corresponding first state estimate and second state estimate.

[0012] When the isolation converter of the auxiliary converter enters the constant gain state, the state estimation equation is constructed on the boost voltage and the intermediate voltage through the transformer gain to obtain the corresponding third state estimate and fourth state estimate.

[0013] After the three-phase inverter of the auxiliary converter is started, the state estimation equation is constructed by modulating the first output voltage, the second output voltage and the intermediate voltage to obtain the corresponding fifth state estimate, sixth state estimate and seventh state estimate.

[0014] When no load is applied to the output of the auxiliary converter, the state estimation equations for the first arm current and the second arm current are constructed using the first output voltage, the second output voltage, and the capacitor to obtain their respective eighth and ninth state estimates.

[0015] After the auxiliary converter is started, a current state estimation equation is constructed for the input voltage, the first output voltage, the second output voltage, the first bridge arm current, and the second bridge arm current by calibrating the operating efficiency, so as to obtain the tenth state estimate.

[0016] On the other hand, the coupling relationships between the sensors in the auxiliary converter are established based on the state estimation equations, including:

[0017] A first coupling relationship between the input voltage and the boost voltage is established based on the duty cycle of the switching devices in the auxiliary converter;

[0018] A second coupling relationship is established between the boost voltage and the intermediate voltage based on the transformer gain;

[0019] A third coupling relationship is established between the first output voltage, the second output voltage, and the intermediate voltage based on the modulation ratio;

[0020] A fourth coupling relationship is established between the first output voltage, the second output voltage, the first bridge arm current, and the second bridge arm current based on the circuit relationship of the auxiliary converter.

[0021] A fifth coupling relationship is established between the input voltage, the first output voltage, the second output voltage, the first bridge arm current, and the second bridge arm current based on the calibrated operating efficiency.

[0022] The final coupling relationship is determined based on the first coupling relationship, the second coupling relationship, the third coupling relationship, the fourth coupling relationship, and the fifth coupling relationship.

[0023] On the other hand, determining the current reference sensor based on the coupling relationship includes:

[0024] If there are adjacent sensors sampling before and after the selected sampling data in the coupling relationship, then the sensor corresponding to the current sampling data in which there are adjacent sensors sampling before and after is taken as the initial reference sensor.

[0025] Among the initial reference sensors, the reference sensor that can be used to calculate the coupling relationship between the sampling data of the preceding and following sensors is selected as the final reference sensor.

[0026] Randomly select one reference sensor from the final reference sensors as the current reference sensor, and designate the remaining sensors other than the current reference sensor as non-reference sensors.

[0027] On the other hand, the fault handling result of each sensor is determined based on the sampling data of the current reference sensor and the state estimate values ​​corresponding to each of the adjacent sensors, and the sampling data of the other sensors besides the current reference sensor and the state estimate values ​​of their corresponding adjacent sensors, including:

[0028] The fault handling result of the current reference sensor is determined by comparing the sampled data of the current reference sensor with the state estimates of the adjacent sensors before and after it.

[0029] When the fault handling result of the current reference sensor is normal, the sensor to be diagnosed after the current reference sensor is determined according to the coupling relationship, and the sensor to be diagnosed is taken as the current sensor to be diagnosed. The fault handling result of the current sensor to be diagnosed is determined by comparing the sampling data of the current sensor to be diagnosed with the state estimate value corresponding to one of the adjacent sensors.

[0030] If the fault handling result of the current sensor to be diagnosed is normal, then the next level sensor of the current sensor to be diagnosed is determined as the new current sensor to be diagnosed according to the coupling relationship, and the process returns to the step of determining the fault handling result of the current sensor to be diagnosed by comparing the sampled data of the current sensor to be diagnosed with the state estimate value of one of the adjacent sensors, until the fault handling of all other sensors is completed.

[0031] On the other hand, the fault handling result of the current reference sensor is determined by comparing the sampled data of the current reference sensor with the state estimates of the adjacent sensors before and after it, including:

[0032] The sampled data of the current reference sensor is compared with the state estimates of the adjacent sensors before and after it.

[0033] If the difference between the sampled data of at least one current reference sensor and the state estimate of each of the adjacent sensors is less than the first threshold, then the fault handling result of the current reference sensor is determined to be normal.

[0034] If the difference between the sampled data of the current reference sensor and the state estimate of each of the adjacent sensors is greater than or equal to the first threshold, then the fault handling result of the current reference sensor is determined to be a fault.

[0035] On the other hand, the fault handling result of the current sensor to be diagnosed is determined by comparing the sampled data of the current sensor to be diagnosed with the state estimate value of one of its adjacent sensors, including:

[0036] If the difference between the sampled data of the current sensor to be diagnosed and the state estimate of one of its adjacent sensors is less than the second threshold, then the fault handling result of the current sensor to be diagnosed is determined to be normal.

[0037] If the difference between the sampled data of the current sensor to be diagnosed and the state estimate of one of its adjacent sensors is greater than or equal to the second threshold, then the fault handling result of the current sensor to be diagnosed is determined to be a fault.

[0038] On the other hand, when the fault handling result of the current reference sensor is determined to be a fault, the method further includes:

[0039] The first difference is obtained by comparing the sampled data of the current reference sensor with the state estimate of the previous adjacent sensor.

[0040] The second difference is obtained by comparing the sampled data of the current reference sensor with the state estimate of the next adjacent sensor;

[0041] If both the first difference and the second difference are less than the first threshold, then the fault handling results of the adjacent sensors are determined to be normal. Then, the state estimation values ​​corresponding to the adjacent sensors are used as the replacement sampling data of the current reference sensor for redundancy.

[0042] On the other hand, if either the first difference or the second difference is greater than or equal to the first threshold, the method further includes:

[0043] If the difference between the state estimates of adjacent sensors is greater than the third threshold, the fault handling result of the adjacent sensors is determined to be a fault, and the system is shut down for protection.

[0044] On the other hand, the state estimates corresponding to each of the preceding and following adjacent sensors are used as replacement sampling data for the current reference sensor, including:

[0045] A first weight parameter and a second weight parameter are preset for the state estimation values ​​corresponding to the adjacent sensors before and after; wherein, the second weight parameter is greater than the first weight parameter;

[0046] The first sampling data is obtained by multiplying the first weight parameter with the state estimate corresponding to the previous adjacent sensor.

[0047] The second weighting parameter is multiplied by the state estimate corresponding to the next adjacent sensor to obtain the second sampling data;

[0048] The first sampled data and the second sampled data are summed to obtain the replacement sampled data of the current reference sensor.

[0049] On the other hand, when the fault handling result of the current sensor to be diagnosed is a fault, the method further includes:

[0050] The state estimate of one of the adjacent sensors corresponding to the current sensor to be diagnosed is used as the replacement sampling data of the current sensor to be diagnosed for redundancy.

[0051] On the other hand, during the sensor redundancy processing, the method further includes:

[0052] If another sensor of the target sensor that is redundantly processed fails, the auxiliary converter will be shut down for protection.

[0053] On the other hand, the method for obtaining the state estimate further includes:

[0054] Obtain historical sampling data;

[0055] Feature parameters are obtained by performing time-series feature processing on the historical sampling data;

[0056] Record the number of calls; call the artificial intelligence model, input the feature parameters into the artificial intelligence model, and output the current state estimate; when the current state estimate is not within the first preset range, increment the number of calls by 1, and return to the step of calling the artificial intelligence model for training, until the current state estimate is within the first preset range or the preset number of calls is reached before the number of calls is reached, thus completing the training process of the artificial intelligence model;

[0057] Correspondingly, the current feature parameters are obtained by performing time-series feature processing on the sampled data, and the trained artificial model is called to input the current feature parameters into the trained artificial model to output the final state estimate.

[0058] To address the aforementioned technical problems, this application also provides a sensor fault handling device, comprising:

[0059] Memory, used to store computer programs;

[0060] A processor is used to implement the steps of the sensor fault handling method as described above when executing the computer program.

[0061] To address the aforementioned technical problems, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the sensor fault handling method described above.

[0062] This application provides a sensor fault handling method. First, it acquires sampling data from each sensor in the auxiliary converter. Based on the sampling data and the corresponding sensor's position in the auxiliary converter, a state estimation equation is constructed to obtain the state estimate value. By quantifying the data mapping relationship of state variables, this application, which can only determine the presence of faults through traditional dual-sensor mutual inspection, facilitates accurate fault location by constructing the state estimation equation and its multi-state variable coupling constraints. Second, it establishes the coupling relationship between each sensor in the auxiliary converter based on the state estimation equation to determine the current reference sensor. Combining the coupling relationship between sensors at different positions in the multi-stage converter of the auxiliary converter, it decouples and determines the state of the reference sensor, establishing the starting point for subsequent forward and backward sequential diagnosis. The faulty sensor in the multi-sensor auxiliary converter system is located through feature logic. Finally, based on the sampling data of the current reference sensor and the state estimate values ​​of each adjacent sensor, and the sampling data of the other sensors besides the current reference sensor and their corresponding state estimate values, the fault handling result of each sensor is determined to complete the fault location processing. This method utilizes the coupling relationship between sensors to select a sensor with adjacent sensors that are simply coupled to each other as the starting point. The reliability of this sensor is then assessed based on the status of the sensors before and after it. Starting from this point, other sensors are diagnosed sequentially. Through specific logic, faulty sensors in the multi-sensor auxiliary transformer system can be located, achieving accurate positioning of multiple sensors. Furthermore, it can operate independently without the need for additional hardware detection or modification of control information, exhibiting high real-time performance.

[0063] In addition, this application also provides a sensor fault handling device and medium, which have the same beneficial effects as the sensor fault handling method described above. Attached Figure Description

[0064] To more clearly illustrate the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0065] Figure 1 A flowchart illustrating a sensor fault handling method provided in this application embodiment;

[0066] Figure 2 A schematic diagram of the circuit topology of an auxiliary converter provided in an embodiment of this application;

[0067] Figure 3 A schematic diagram illustrating the establishment of a coupling relationship provided in an embodiment of this application;

[0068] Figure 4 This application provides a schematic diagram of sensor fault diagnosis.

[0069] Figure 5 A schematic diagram illustrating redundant operation and shutdown protection for sensor fault diagnosis provided in an embodiment of this application;

[0070] Figure 6 A structural diagram of a sensor fault handling device provided in an embodiment of this application;

[0071] Figure 7 A structural diagram of a sensor fault handling device provided in an embodiment of this application;

[0072] Figure 8 A flowchart illustrating sensor redundancy protection provided in this application embodiment;

[0073] Figure 9 This is a flowchart illustrating a sensor redundancy control method provided in an embodiment of this application. Detailed Implementation

[0074] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of this application.

[0075] The core of this application is to provide a sensor fault handling method, device, and medium to solve the problem of positioning difficulties and reduced reliability of auxiliary converters caused by the use of dual sensor mutual detection in conventional solutions.

[0076] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0077] With the continuous development of high-speed rail, people's requirements for travel comfort are also increasing. As the interface between the power grid and on-board equipment, converters play an important role and have therefore experienced rapid development in recent years.

[0078] Sensors are critical components for the normal and reliable operation of converters. A typical auxiliary converter system includes several or even dozens of current, voltage, and other sensors that collect status information from the converter's input, intermediate, and output levels. This information is then used to control and operate active devices such as switches within the converter. Furthermore, sensors can be used to diagnose system operating conditions and provide timely protection. However, sensor failure can cause the system's control to deviate from its normal state and can also lead to the failure of system protection mechanisms. Therefore, it is necessary to diagnose the sensors in the converter system to identify faulty sensors in a timely manner.

[0079] Auxiliary converters typically have a multi-stage structure and a large number of sensors with interconnected data acquisition states. Current sensor diagnostics often rely on cross-verification between adjacent sensors. This involves calculating an estimated value for sensor B from the sampled value of sensor A and comparing it with the sampled value of sensor B. The degree of deviation is then used to determine if sensor B has failed. However, this state estimation is reciprocal, simultaneously using the sampled value of sensor B to diagnose sensor A. Therefore, when sensor B fails, both sensor A and B may be reported as faulty simultaneously, making it difficult to accurately locate the faulty sensor in practical applications. The sensor fault handling method provided in this application solves these problems.

[0080] Figure 1 A flowchart of a sensor fault handling method provided in an embodiment of this application is shown below. Figure 1 As shown, the method includes:

[0081] S11: Acquire sampling data from each sensor of the auxiliary converter;

[0082] S12: Construct state estimation equations based on the sampling data and the corresponding sensor positions in the auxiliary converter to obtain state estimates;

[0083] S13: Establish the coupling relationship between the sensors in the auxiliary converter based on the state estimation equations to determine the current reference sensor;

[0084] S14: In the coupling relationship, the fault handling result of each sensor is determined based on the sampling data of the current reference sensor and the state estimate value of each adjacent sensor, and the sampling data of the other sensors besides the current reference sensor and the state estimate value of each corresponding adjacent sensor, so as to complete the fault location processing.

[0085] Specifically, Figure 2 A schematic diagram of the circuit topology of an auxiliary converter provided in this application embodiment is shown below. Figure 2As shown, the overall circuit is divided into three stages. The DC grid voltage is input, regulated by a first-stage boost converter, then connected to a first-stage constant-gain isolated DC-DC (DCDC) converter to achieve electrical isolation and step down the voltage, and finally a first-stage three-phase inverter to achieve DC-AC conversion and power supply to the subsequent vehicle auxiliary equipment.

[0086] Will Figure 2 The system collects sampling data from various sensors within the auxiliary converter. State estimation equations are constructed using these data and the corresponding sensor locations on the auxiliary converter to obtain state estimates. It should be noted that these state estimation equations can be constructed using either a circuit model or a mathematical model; no specific limitation is made here. Overall, the system overcomes the limitations of direct sensor measurement by quantifying the mathematical mapping relationship between state variables, achieving accurate perception of core states, system performance optimization, and improved reliability throughout the entire lifecycle. Due to physical characteristics and installation conditions, some key state variables in the auxiliary converter are limited; the state estimation equations achieve quantitative estimation of these states through the coupling mapping between measurable sensor data and state variables.

[0087] Specifically, state estimates are constructed by considering different states of the auxiliary converter, such as the continuous state of the boost converter, the constant gain state of the isolated DC-DC converter, the state corresponding to the small inductor voltage drop after the three-phase inverter starts up, the state before the output is connected to the load, and the state of the input current after the converter starts up. State estimation equations are then constructed based on these conditions.

[0088] In some embodiments, the sampled data includes at least input voltage, input current, boost voltage, intermediate voltage, first bridge arm current, second bridge arm current, first output voltage, and second output voltage; a state estimation equation is constructed based on each sampled data and the corresponding sensor location on the auxiliary converter to obtain a state estimate, including:

[0089] When the boost converter of the auxiliary converter enters the continuous state, the state estimation equation is constructed based on the duty cycle of the switching devices in the auxiliary converter to obtain the corresponding first state estimate and second state estimate.

[0090] When the isolation converter of the auxiliary converter enters the constant gain state, the state estimation equation is constructed on the boost voltage and intermediate voltage through the transformer gain to obtain the corresponding third state estimate and fourth state estimate.

[0091] After the three-phase inverter of the auxiliary converter is started, the state estimation equation is constructed by modulating the first output voltage, the second output voltage and the intermediate voltage to obtain the corresponding fifth state estimate, sixth state estimate and seventh state estimate.

[0092] When no load is applied to the output of the auxiliary converter, the state estimation equations for the first bridge arm current and the second bridge arm current are constructed using the first output voltage, the second output voltage, and the capacitor to obtain their respective eighth and ninth state estimates.

[0093] After the auxiliary converter is started, the current state estimation equation is constructed by calibrating the operating efficiency for the input voltage, the first output voltage, the second output voltage, the first bridge arm current and the second bridge arm current, so as to obtain the tenth state estimate.

[0094] Specifically, the main sensors in the auxiliary transformer system include: input voltage. Input current Boost voltage Intermediate voltage First arm current Second arm current First output voltage Second output voltage .

[0095] (1) The boost converter enters continuous operation, and the input voltage... With boost voltage Duty cycle of switching devices Mutual calculation yields the first state estimate. Second state estimate ;

[0096] ;

[0097] .

[0098] (2) The isolated DC-DC converter enters the constant gain state, and the boost voltage... With intermediate voltage Inter-transformer gain Mutual calculation yields the third state estimate. and fourth state estimate ;

[0099] ;

[0100] .

[0101] (3) After the three-phase inverter starts, the inductor voltage drop is small, and the first output voltage is low. Second output voltage With intermediate voltage Inter-modulation ratio By performing mutual calculations, the fifth state estimate is obtained. Sixth state estimate and the seventh state estimate ;

[0102] ;

[0103] ;

[0104] in, The first output voltage The corresponding modulation ratio, For the second output voltage The corresponding modulation ratio.

[0105] (4) The first bridge arm current before the output is connected to the load. Second arm current It can be achieved through the first output voltage Second output voltage With capacitance parameters The calculation yields the estimated value of the eighth state. and the ninth state estimate ;

[0106] ;

[0107] .

[0108] (5) Input current after the converter starts up Based on input voltage First output voltage Second output voltage First arm current Second arm current Compared with the calibrated operating efficiency Together, we calculated and obtained the estimate of the tenth state. .

[0109] .

[0110] This embodiment provides a state estimation equation based on a circuit model to obtain a state estimate. It transforms the converter's topological constraints, energy conservation laws, and component electrical characteristics into quantifiable mathematical mappings. The unmeasurable state is reconstructed using measurable sensor data and circuit physical constraints. The state estimation equation of the circuit model not only provides accurate state feedback but also predicts state change trends based on physical laws, offering forward-looking data for control optimization and fault protection.

[0111] In other embodiments, the process of obtaining the state estimate further includes:

[0112] Obtain historical sampling data;

[0113] Feature parameters are obtained by performing time-series feature processing on historical sampling data;

[0114] Record the number of calls; call the artificial intelligence model, input the feature parameters into the artificial intelligence model, and output the current state estimate; if the current state estimate is not within the first preset range, increment the number of calls by 1, and return to the step of calling the artificial intelligence model for training, until the current state estimate is within the first preset range before the number of calls is reached or the preset number of calls is reached, thus completing the training process of the artificial intelligence model;

[0115] Correspondingly, the current feature parameters are obtained by performing time-series feature processing on the sampled data, and the trained artificial model is called to input the current feature parameters into the trained artificial model to output the final state estimate.

[0116] Specifically, the artificial intelligence model here can be a machine learning model or a deep learning model, which can accurately estimate key quantities such as inductor current, capacitor voltage, and the health status of switching transistors through the mapping relationship between sensor data and state variables.

[0117] The sensor's sampling data needs to cover the entire operating conditions of the converter, including electrical quantities (corresponding sampling data), physical quantities (heat sink temperature and ambient humidity), and time-series characteristics (maximum / minimum / mean / variance within the sliding window) for quantification constraints.

[0118] Sensor outliers are removed, input features are normalized, and time-series aligned.

[0119] Pearson correlation coefficient can be used to screen features that are strongly correlated with state variables, or redundant features can be eliminated through recursive feature elimination to reduce model complexity.

[0120] If deep learning is used, the network structure is as follows: Input layer (feature dimension n) → Hidden layer (2-3 layers, 64-128 neurons per layer, activation function (Rectified Linear Unit, ReLU)) → Output layer (state variable dimension n) The mean squared error loss is determined through a loss function for application. A long short-term memory (LSTM) network is employed, using gating units (input gate, forget gate, output gate) to capture the long-short-term dependencies of time-series data, making it suitable for handling dynamic changes within the converter switching cycle (such as the transient process of inductor current). Quantization structure: Input: Time-series feature sequence (length...) (i.e., sensor data from 20 consecutive sampling points); Network layers: Long Short-Term Memory Layer (LSTM) (2 layers, 128 neurons per layer) → Dropout layer (dropout rate=0.2, to prevent overfitting) → Fully connected layer (outputs state estimates); The loss function uses a hybrid loss function. If a hybrid model is used, Convolutional Neural Networks (CNN) extracts local waveform features, and LSTM captures temporal dependencies. Combining the advantages of both, it is suitable for complex scenarios with strong nonlinearity and temporal dynamics; Quantization structure: CNN layer (extracts waveform features) → LSTM layer (captures temporal correlations) → Output layer.

[0121] This embodiment provides a method for constructing state estimation equations using mathematical models. By employing a data-driven approach, it avoids the reliance of traditional models on precise mathematical modeling. At the same time, it ensures that the model meets actual engineering needs and improves positioning accuracy by using quantitative indicators (error, delay, robustness).

[0122] After obtaining the state estimate in step S12, step S13 establishes the coupling relationship between the sensors in the auxiliary converter based on the state estimate equations to determine the current reference sensor. This coupling relationship involves first determining the core topology of the converter, such as the rectifier-filter-inverter structure commonly used in rail transit, and identifying key energy storage components (inductors and capacitors). Their current and voltage are the core state variables in the state equations. Simultaneously, the sensor configuration and detection objects are clarified. The state variables and input / output variables in the state equations mostly correspond to the sensor detection quantities. Through elimination, transformation, and other operations, the coupling expression between sensors can be obtained. After derivation, the accuracy of the coupling relationship needs to be verified. On the one hand, a converter simulation model can be built using Matlab / Simulink to compare the measured sensor data with the calculated results of the coupling equations. For example, changing the load size can be used to observe whether the changes in the detection quantities of the current and voltage sensors conform to the derived coupling rules. On the other hand, the influence of parasitic parameters must be considered. Parasitic capacitance and lead inductance of the power module can introduce additional coupling paths, requiring correction of the parameters in the state equations to ensure that the sensor coupling relationship accurately reflects the actual circuit characteristics.

[0123] In some embodiments, establishing the coupling relationship between the sensors in the auxiliary converter based on the state estimation equations includes:

[0124] The first coupling relationship between the input voltage and the boost voltage is established based on the duty cycle of the switching devices in the auxiliary converter;

[0125] A second coupling relationship is established between the step-up voltage and the intermediate voltage based on the transformer gain;

[0126] A third coupling relationship is established between the first output voltage, the second output voltage, and the intermediate voltage based on the modulation ratio;

[0127] A fourth coupling relationship is established between the first output voltage, the second output voltage, the first bridge arm current, and the second bridge arm current based on the circuit relationship of the auxiliary converter.

[0128] A fifth coupling relationship is established between the input voltage, the first output voltage, the second output voltage, the first bridge arm current, and the second bridge arm current based on the calibrated operating efficiency.

[0129] The final coupling relationship is determined based on the first coupling relationship, the second coupling relationship, the third coupling relationship, the fourth coupling relationship, and the fifth coupling relationship.

[0130] Specifically, Figure 3 This is a schematic diagram illustrating the establishment of a coupling relationship in an embodiment of this application, as shown below. Figure 3 As shown, a first coupling relationship between the input voltage and the boost voltage is established based on the duty cycle of the switching devices in the auxiliary converter; a second coupling relationship between the boost voltage and the intermediate voltage is established based on the transformer gain; a third coupling relationship between the first output voltage, the second output voltage, and the intermediate voltage is established based on the modulation ratio; a fourth coupling relationship between the first output voltage, the second output voltage, the first bridge arm current, and the second bridge arm current is established based on the circuit relationship of the auxiliary converter; and a fifth coupling relationship between the input voltage, the first output voltage, the second output voltage, the first bridge arm current, and the second bridge arm current is established based on the calibrated operating efficiency.

[0131] This embodiment establishes the coupling relationship between various sensors of the auxiliary converter based on the state equation. Through quantitative mapping such as circuit topology law and energy conservation, the dispersed sensor data is connected in series into a network using circuit physical laws. This not only solves the problems of unreliable data, incomplete coverage, and high cost of traditional independent sensor operation, but also provides a quantitative mathematical basis for the precise control, fault protection, and low-cost operation and maintenance of the auxiliary converter, which is convenient for subsequent fault location and processing.

[0132] The current reference sensor is determined by coupling relationship. This involves selecting a sensor that is adjacent to the previous and next sensors and can be easily coupled, as the starting point. The reliability of the sensor is first diagnosed based on the status of the previous and next sensors. Then, other sensors are diagnosed sequentially from the previous and next sensors. Through specific logic, the faulty sensor in the multi-sensor auxiliary transformer system can be located.

[0133] In some embodiments, determining the current reference sensor based on the coupling relationship includes:

[0134] If there are adjacent sensors sampling before and after the selected sampling data in the coupling relationship, then the sensor corresponding to the current sampling data in which there are adjacent sensors sampling before and after is taken as the initial reference sensor.

[0135] Among the initial reference sensors, the reference sensor that can be used to calculate the coupling relationship between the sampling data of the preceding and following sensors is selected as the final reference sensor.

[0136] Randomly select one reference sensor from the final reference sensors as the current reference sensor, and designate the remaining sensors other than the current reference sensor as non-reference sensors.

[0137] Specifically, such as Figure 3 As shown, the sampling data corresponding to the sensors can be determined to be adjacent in the coupling relationship (the order from left to right determines the sensors corresponding to the sampling data). In the coupling relationship, the sampling data is filtered to find that there are adjacent sensors sampling before and after it. Figure 3 The middle can be a boost voltage Intermediate voltage First arm current Second arm current First output voltage Second output voltage .

[0138] Considering that the coupling relationship can be calculated at both the front and rear, then in the above-mentioned boost voltage... Intermediate voltage First arm current Second arm current First output voltage Second output voltage After screening, the final reference sensor (boost voltage) is obtained. Intermediate voltage ).

[0139] Choose any one of the final reference sensors as the current reference sensor, and update the remaining reference sensors to non-reference sensors, merging them into the subsequent sensors to be diagnosed.

[0140] The current reference sensor determination process provided in this embodiment diagnoses the reliability of the sensor based on the status of the preceding and following sensors, and uses this as a starting point to diagnose other sensors sequentially forward and backward. Through specific logic, the faulty sensor in the multi-sensor auxiliary transformer system can be located, ensuring the orderly handling of faults in multiple sensors.

[0141] In step S14, within the coupling relationship, the sampling data of the current reference sensor is first compared with the state estimates of adjacent sensors. If the comparison is successful, the current reference sensor is determined to be fault-free, and the process continues with the sampling data of the remaining sensors and their respective state estimates to determine fault handling. Since there are multiple remaining sensors, the current reference sensor is located based on the topology recorded in the coupling relationship, and then the remaining sensors are checked and processed sequentially. It should be noted that only one sensor is allowed to fail in the entire scheme; if multiple sensors fail, a shutdown protection process will be implemented.

[0142] This application provides a sensor fault handling method. First, it acquires sampling data from each sensor in an auxiliary converter. Based on the sampling data and the corresponding sensor's position in the auxiliary converter, a state estimation equation is constructed to obtain a state estimate. By quantifying the data mapping relationship of state variables, this application, which can only determine the presence of a fault through traditional dual-sensor mutual inspection, uses the construction of state estimation equations and the coupling constraints of multiple state variables to facilitate accurate fault location. Second, it establishes the coupling relationship between each sensor in the auxiliary converter based on the state estimation equations to determine the current reference sensor. Combining the coupling relationship between sensors at different positions in the multi-stage converter of the auxiliary converter, it decouples and determines the state of the reference sensor, establishing a starting point for subsequent forward and backward sequential diagnosis. The faulty sensor in the multi-sensor auxiliary converter system is located through feature logic. Finally, based on the sampling data of the current reference sensor and the state estimates of each adjacent sensor, and the sampling data of the other sensors besides the current reference sensor and the state estimates of their corresponding adjacent sensors, the fault handling result of each sensor is determined to complete the fault location process. This method utilizes the coupling relationship between sensors to select a sensor with adjacent sensors that are simply coupled to each other as the starting point. The reliability of this sensor is then assessed based on the status of the sensors before and after it. Starting from this point, other sensors are diagnosed sequentially. Through specific logic, faulty sensors in the multi-sensor auxiliary transformer system can be located, achieving accurate positioning of multiple sensors. Furthermore, it can operate independently without the need for additional hardware detection or modification of control information, exhibiting high real-time performance.

[0143] In some embodiments, the fault handling result of each sensor is determined based on the sampling data of the current reference sensor and the state estimates corresponding to each of the adjacent sensors, and the sampling data of the other sensors besides the current reference sensor and the state estimates corresponding to their respective adjacent sensors, including:

[0144] The fault handling result of the current reference sensor is determined by comparing the sampled data of the current reference sensor with the state estimates of the adjacent sensors before and after it.

[0145] When the fault handling result of the current reference sensor is normal, the sensor to be diagnosed after the current reference sensor is determined according to the coupling relationship, and the sensor to be diagnosed is taken as the current sensor to be diagnosed. The fault handling result of the current sensor to be diagnosed is determined by comparing the sampling data of the current sensor to be diagnosed with the state estimate value of one of the adjacent sensors.

[0146] If the fault handling result of the current sensor to be diagnosed is normal, then the next level sensor of the current sensor to be diagnosed is determined as the new current sensor to be diagnosed based on the coupling relationship, and the process returns to the step of comparing the sampled data of the current sensor to be diagnosed with the state estimate value of one of the adjacent sensors to determine the fault handling result of the current sensor to be diagnosed, until the fault handling of all other sensors is completed.

[0147] Specifically, the sampling data of the current reference sensor is compared with the state estimates of its adjacent sensors. This comparison is performed separately. Only when the fault handling result of the current reference sensor is normal can the comparison processing of the subsequent sensor to be diagnosed proceed. Here, the sensor to be diagnosed after the current reference sensor needs to be determined based on the coupling relationship, and this sensor is designated as the current sensor to be diagnosed. The sampling data of the current sensor to be diagnosed is compared with the state estimate of one of its adjacent sensors to determine the fault handling result of the current sensor to be diagnosed. Only when the fault handling result of the current sensor to be diagnosed is normal can the next sensor to be diagnosed be obtained as the new current sensor to be diagnosed. This process continues until the faults of all sensors have been diagnosed.

[0148] Only when the current reference sensor is functioning correctly can subsequent diagnostics of other sensors be performed, and so on, diagnosing sensors sequentially based on their coupling relationships. If there are no subsequent related sensors, fault handling can be performed in parallel with sensors of the same level.

[0149] The parallel processing method provided in this embodiment for sensors at the same level under the coupling relationship can be implemented. For sensors at different levels, the fault processing of the next level sensor can only be performed if the fault processing result of the previous level sensor is normal. This ensures the orderly fault processing of the sensors and saves fault diagnosis efficiency.

[0150] In some embodiments, determining the fault handling result of the current reference sensor by comparing the sampled data of the current reference sensor with the state estimates of the adjacent sensors before and after it includes:

[0151] The sampled data from the current reference sensor is compared with the state estimates of the adjacent sensors before and after it.

[0152] If the difference between the sampled data of at least one current reference sensor and the state estimate of each of the preceding and following adjacent sensors is less than the first threshold, then the fault handling result of the current reference sensor is determined to be normal.

[0153] If the difference between the sampled data of the current reference sensor and the state estimate of each of the adjacent sensors is greater than or equal to the first threshold, then the fault handling result of the current reference sensor is determined to be a fault.

[0154] Specifically, if the difference between the sampled data of at least one current reference sensor and the state estimate of each of the preceding and following adjacent sensors is less than a first threshold, then the fault handling result of the current reference sensor is determined to be normal; otherwise, the fault handling result of the current reference sensor is determined to be faulty.

[0155] Figure 4 This application provides a schematic diagram of a sensor fault diagnosis, as shown in the embodiment. Figure 4 As shown, using a boost voltage sensor, the second state estimate is calculated from two estimated values ​​obtained by the front-end and rear-end sensors. and third state estimate respectively with boost voltage The comparison revealed the following:

[0156] (1) When both comparisons are successful, the sensor is considered to be normal;

[0157] (2) If one comparison fails and another comparison succeeds, the sensor is considered to be normal;

[0158] (3) If the two comparisons fail, the sensor is deemed to be faulty.

[0159] The fault handling result determination process of the current reference sensor provided in this embodiment improves the fault accuracy of the current reference sensor by comparing the state estimates and sampled data of adjacent sensors. The state estimates (including circuit physical constraints) derived from the state equations of adjacent time steps are compared with the current sensor sampled data in a spatiotemporal dimension, replacing traditional single-point threshold judgment or static dual-sensor mutual inspection. This solves the problems of weak anti-interference, high false positive rate, and incomplete coverage of traditional methods, without requiring additional hardware costs. Furthermore, it extends the sensor's lifespan and reduces maintenance costs through latent fault early warning.

[0160] In some embodiments, the fault handling result of the current sensor to be diagnosed is determined by comparing the sampled data of the current sensor to be diagnosed with the state estimate value of one of its adjacent sensors, including:

[0161] If the difference between the sampled data of the sensor to be diagnosed and the state estimate of one of the adjacent sensors is less than the second threshold, then the fault handling result of the sensor to be diagnosed is determined to be normal.

[0162] If the difference between the sampled data of the sensor to be diagnosed and the state estimate of one of its adjacent sensors is greater than or equal to the second threshold, then the fault handling result of the sensor to be diagnosed is determined to be a fault.

[0163] Specifically, if the difference between the sampled data of the current sensor to be diagnosed and the state estimate of one of its adjacent sensors is less than the second threshold, then the fault handling result of the current sensor to be diagnosed is determined to be normal; otherwise, it is a fault.

[0164] like Figure 3 As shown, taking the intermediate voltage sensor as an example, when the boost voltage sensor determines that it is working normally, the fourth state estimate value calculated by the boost voltage sensor will be used. With intermediate voltage The comparison revealed the following:

[0165] (1) When the comparison is successful, the sensor is deemed to be normal;

[0166] (2) When the comparison fails, the sensor is determined to be faulty.

[0167] If the intermediate voltage sensor is normal, continue to check the output voltage sensor, and then proceed to diagnose the faults of the next sensor in sequence. This will not be elaborated further.

[0168] The fault handling method for the sensor to be diagnosed provided in this embodiment achieves accurate fault identification and type differentiation by combining the temporal correlation of the current sample and the single adjacent state estimate with the physical constraints of the state equation. Through the temporal constraints of the single adjacent state estimate and the physical laws of the state equation, the detection reliability in complex scenarios is significantly improved.

[0169] In conventional solutions, a failure in a converter sensor causes the converter to malfunction and shut down for protection, preventing continued operation and reducing the lifespan of subsequent sensors. In some embodiments, when the fault handling result of the current reference sensor is determined to be a fault, the method further includes:

[0170] The first difference is obtained by comparing the sampled data of the current reference sensor with the state estimate of the previous adjacent sensor.

[0171] The second difference is obtained by comparing the sampled data of the current reference sensor with the state estimate of the next adjacent sensor;

[0172] If both the first difference and the second difference are less than the first threshold, then the fault handling results of the adjacent sensors are determined to be normal. Then, the state estimates corresponding to the adjacent sensors are used as the replacement sampling data of the current reference sensor for redundancy.

[0173] Specifically, when the fault result of the current reference sensor is determined to be faulty, the state estimates of adjacent sensors are compared again. The sampling data of the current reference sensor is compared with the state estimate of the previous adjacent sensor to obtain a first difference; the sampling data of the current reference sensor is compared with the state estimate of the next adjacent sensor to obtain a second difference. The two differences are compared with a first threshold. If they are less than the threshold, it indicates that the previous and next adjacent sensors are normal. In this case, the state estimates of the previous and next adjacent sensors are used to replace the sampling data of the faulty sensor for redundancy.

[0174] In this embodiment, when the current reference sensor fails, the state estimate obtained through state calculation replaces the sampled value of the failed sensor, enabling redundant operation of the converter. This enhances the converter's reliability and emergency response capabilities. It ensures continuous system operation while extending the lifespan of other sensors. It prevents the system from falling into control malfunctions or shutting down due to sensor failure. The state estimate and the sampled data before the fault are based on the same physical model, allowing for seamless and shock-free switching.

[0175] In some embodiments, if the first difference and the second difference are greater than or equal to a first threshold, the method further includes:

[0176] If the difference between the state estimates of adjacent sensors is greater than the third threshold, the fault handling result of the adjacent sensors is determined to be a fault, and the system is shut down for protection.

[0177] Specifically, if the first difference and the second difference in the state estimates of the sensors used for calibration are greater than or equal to the first threshold, i.e. the comparison fails, it is determined that there is a fault in the adjacent sensors and a shutdown protection process is required.

[0178] This embodiment provides a timely shutdown protection process in case of failure of adjacent sensors during redundant operation, ensuring the reliability of the system.

[0179] In some embodiments, the state estimates corresponding to each of the preceding and following adjacent sensors are used as replacement sampling data for the current reference sensor, including:

[0180] A first weight parameter and a second weight parameter are pre-set for the state estimates of each adjacent sensor; wherein the second weight parameter is greater than the first weight parameter.

[0181] The first sampling data is obtained by multiplying the first weighting parameter with the state estimate corresponding to the previous adjacent sensor.

[0182] The second sampling data is obtained by multiplying the second weighting parameter with the state estimate corresponding to the next adjacent sensor.

[0183] The first sampled data and the second sampled data are summed to obtain the replacement sampled data of the current reference sensor.

[0184] Specifically, such as Figure 2 As shown, the replacement sampling data is obtained by weighting the first weight parameter and the second weight parameter of the state estimate values ​​of the adjacent sensors before and after the current reference sensor with the corresponding state estimate values.

[0185] The reason the second weighting parameter is greater than the first weighting parameter is that the open-loop gain of the state estimate under the second weighting parameter is a fixed value, which is relatively stable. The input voltage variation of the first weighting parameter is unstable and has low accuracy; therefore, the second weighting parameter will be greater than the first weighting parameter.

[0186] This embodiment provides a method to obtain replacement sampling data by weighting the state estimates of adjacent sensors with corresponding weight parameters to achieve redundancy, thereby improving the stability and reliability of the system.

[0187] In some embodiments, when the fault handling result of the current sensor to be diagnosed is a fault, the method further includes:

[0188] The state estimate of one of the adjacent sensors corresponding to the sensor currently under diagnosis is used as the replacement sampling data of the sensor currently under diagnosis for redundancy.

[0189] Specifically, after a fault occurs in the current sensor to be diagnosed, the state estimate of one of the adjacent sensors used for diagnosis needs to be used as replacement sampling data for the current sensor to be diagnosed, in order to perform redundant work.

[0190] The redundancy provided in this embodiment after the current sensor to be diagnosed fails extends the service life of subsequent sensors and reduces the number of times faulty sensors need to be replaced.

[0191] In some embodiments, during sensor redundancy processing, the method further includes:

[0192] If another sensor in the target sensor that is redundantly processed fails, the auxiliary converter will be shut down for protection.

[0193] Specifically, Figure 5 A schematic diagram illustrating redundant operation and shutdown protection for sensor fault diagnosis provided in this application embodiment is shown below. Figure 5 As shown, only one sensor is allowed to fail here. If another sensor fails after redundancy is implemented, the system needs to be shut down.

[0194] This embodiment allows for use in situations where a second sensor failure is not permitted, ensuring system reliability while also improving the accuracy of fault location.

[0195] The foregoing has described in detail various embodiments of the sensor fault handling method. Based on this, this application also discloses a sensor fault handling device corresponding to the above method. Figure 6 This is a structural diagram of a sensor fault handling device provided in an embodiment of this application. Figure 6 As shown, the sensor fault handling equipment includes:

[0196] The acquisition module 11 is used to acquire the sampling data of each sensor of the auxiliary converter;

[0197] Module 12 is used to construct state estimation equations based on the sampling data and the corresponding sensor positions in the auxiliary converter to obtain state estimation values.

[0198] Module 13 is used to establish the coupling relationship between the sensors in the auxiliary converter based on the state estimation equations, so as to determine the current reference sensor;

[0199] The processing module 14 is used to determine the fault handling result of each sensor in the coupling relationship based on the sampling data of the current reference sensor and the state estimate value of each adjacent sensor, and the sampling data of the other sensors besides the current reference sensor and the state estimate value of each corresponding adjacent sensor, so as to complete the fault location processing.

[0200] Since the embodiments of the device part correspond to the embodiments described above, please refer to the embodiments of the method part for the description of the device part, and will not be repeated here.

[0201] For a description of the sensor fault handling device provided in this application, please refer to the above method embodiments. This application will not repeat the description here, but it has the same beneficial effects as the above sensor fault handling method.

[0202] Figure 7 A structural diagram of a sensor fault handling device provided in an embodiment of this application is shown below. Figure 7 As shown, the device includes:

[0203] Memory 21 is used to store computer programs;

[0204] Processor 22 is used to implement the steps of a sensor fault handling method when executing a computer program.

[0205] The sensor fault handling device provided in this embodiment may include, but is not limited to, tablet computers, laptop computers, or desktop computers.

[0206] The processor 22 may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor 22 may be implemented using at least one of the following hardware forms: Digital Signal Processor (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 22 may also include a main processor and a coprocessor. The main processor, also known as the Central Processing Unit (CPU), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor 22 may integrate a Graphics Processing Unit (GPU), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, the processor 22 may also include an Artificial Intelligence (AI) processor, which handles computational operations related to machine learning.

[0207] The memory 21 may include one or more computer-readable storage media, which may be non-transitory. The memory 21 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In this embodiment, the memory 21 is used to store at least the following computer program 211, which, after being loaded and executed by the processor 22, is capable of implementing the relevant steps of the sensor fault handling method disclosed in any of the foregoing embodiments. In addition, the resources stored in the memory 21 may also include an operating system 212 and data 213, etc., and the storage method may be temporary storage or permanent storage. The operating system 212 may include Windows, Unix, Linux, etc. The data 213 may include, but is not limited to, the data involved in the sensor fault handling method, etc.

[0208] In some embodiments, the sensor fault handling device may further include a display screen 23, an input / output interface 24, a communication interface 25, a power supply 26, and a communication bus 27.

[0209] Those skilled in the field can understand, Figure 7 The structure shown does not constitute a limitation on the sensor fault handling device and may include more or fewer components than shown.

[0210] The processor 22 implements the sensor fault handling method provided in any of the above embodiments by calling the instructions stored in the memory 21.

[0211] For a description of the sensor fault handling device provided in this application, please refer to the above method embodiments. This application will not repeat the description here, but it has the same beneficial effects as the above sensor fault handling method.

[0212] Furthermore, this application also provides a computer-readable storage medium storing a computer program, which, when executed by processor 22, implements the steps of the sensor fault handling method described above.

[0213] It is understood that if the methods in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and executes all or part of the steps of the methods in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0214] For a description of the computer-readable storage medium provided in this application, please refer to the above method embodiments. This application will not repeat the description here, but it has the same beneficial effects as the above sensor fault handling method.

[0215] Figure 8 A flowchart illustrating sensor redundancy protection provided in this application embodiment is shown below. Figure 8 As shown, it includes:

[0216] S21: Sensor A obtains a sampled value;

[0217] S22: Calculate the estimated value of sensor A;

[0218] S23: Determine whether sensor A is faulty based on the sampled value and the estimated value. If yes, proceed to step S24; otherwise, proceed to step S25.

[0219] S24: Estimated value replaces sampled value;

[0220] S25: Determine whether the converter is under protection. If yes, proceed to step S26; otherwise, return to step S21.

[0221] S26: Shutdown protection.

[0222] Figure 9 A flowchart of a sensor redundancy control provided in an embodiment of this application is shown below. Figure 9 As shown, it includes:

[0223] S21: Sensor A obtains a sampled value;

[0224] S22: Calculate the estimated value of sensor A;

[0225] S23: Determine whether sensor A is faulty based on the sampled value and the estimated value. If yes, proceed to step S24; otherwise, proceed to step S27.

[0226] S24: Estimated value replaces sampled value;

[0227] S27: Data processing;

[0228] S28: Closed-loop control, return to step S21.

[0229] The foregoing has provided a detailed description of a sensor fault handling method, apparatus, and medium provided in this application. The various embodiments in the specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section. It should be noted that those skilled in the art can make several improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of this application.

[0230] It should also be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

Claims

1. A sensor fault handling method, characterized in that, include: Acquire sampling data from each sensor of the auxiliary converter; Based on the sampling data and the corresponding sensor positions in the auxiliary converter, a state estimation equation is constructed to obtain the state estimate value; The coupling relationship between the sensors in the auxiliary converter is established based on the state estimation equations to determine the current reference sensor. In the coupling relationship, the fault handling result of each sensor is determined based on the sampling data of the current reference sensor and the state estimation value of each adjacent sensor, and the sampling data of the other sensors besides the current reference sensor and the state estimation value of each corresponding adjacent sensor, so as to complete the fault location processing. Correspondingly, the sampling data includes at least the input voltage, input current, boost voltage, intermediate voltage, first bridge arm current, second bridge arm current, first output voltage, and second output voltage; Based on the sampled data and the corresponding sensor locations at the auxiliary converter, a state estimation equation is constructed to obtain the state estimate, including: When the boost converter of the auxiliary converter enters the continuous state, the state estimation equation is constructed based on the duty cycle of the switching devices in the auxiliary converter to obtain the corresponding first state estimate and second state estimate. When the isolation converter of the auxiliary converter enters the constant gain state, the state estimation equation is constructed on the boost voltage and the intermediate voltage through the transformer gain to obtain the corresponding third state estimate and fourth state estimate. After the three-phase inverter of the auxiliary converter is started, the state estimation equation is constructed by modulating the first output voltage, the second output voltage and the intermediate voltage to obtain the corresponding fifth state estimate, sixth state estimate and seventh state estimate. When no load is applied to the output of the auxiliary converter, the state estimation equations for the first arm current and the second arm current are constructed using the first output voltage, the second output voltage, and the capacitor to obtain their respective eighth and ninth state estimates. After the auxiliary converter is started, a current state estimation equation is constructed for the input voltage, the first output voltage, the second output voltage, the first bridge arm current, and the second bridge arm current by calibrating the operating efficiency, so as to obtain the tenth state estimate. Correspondingly, the coupling relationships between the sensors in the auxiliary converter are established based on the state estimation equations, including: A first coupling relationship between the input voltage and the boost voltage is established based on the duty cycle of the switching devices in the auxiliary converter; A second coupling relationship is established between the boost voltage and the intermediate voltage based on the transformer gain; A third coupling relationship is established between the first output voltage, the second output voltage, and the intermediate voltage based on the modulation ratio; A fourth coupling relationship is established between the first output voltage, the second output voltage, the first bridge arm current, and the second bridge arm current based on the circuit relationship of the auxiliary converter. A fifth coupling relationship is established between the input voltage, the first output voltage, the second output voltage, the first bridge arm current, and the second bridge arm current based on the calibrated operating efficiency. The final coupling relationship is determined based on the first coupling relationship, the second coupling relationship, the third coupling relationship, the fourth coupling relationship, and the fifth coupling relationship; Correspondingly, determining the current reference sensor based on the coupling relationship includes: If there are adjacent sensors sampling before and after the selected sampling data in the coupling relationship, then the sensor corresponding to the current sampling data in which there are adjacent sensors sampling before and after is taken as the initial reference sensor. Among the initial reference sensors, the reference sensor that can be used to calculate the coupling relationship between the sampling data of the preceding and following sensors is selected as the final reference sensor. Randomly select one reference sensor from the final reference sensors as the current reference sensor, and designate the remaining sensors other than the current reference sensor as non-reference sensors.

2. The sensor fault handling method according to claim 1, characterized in that, The fault handling results for each sensor are determined based on the sampling data of the current reference sensor and the state estimates of each of the adjacent sensors, as well as the sampling data of the other sensors besides the current reference sensor and the state estimates of their corresponding adjacent sensors. This includes: The fault handling result of the current reference sensor is determined by comparing the sampled data of the current reference sensor with the state estimates of the adjacent sensors before and after it. When the fault handling result of the current reference sensor is normal, the sensor to be diagnosed after the current reference sensor is determined according to the coupling relationship, and the sensor to be diagnosed is taken as the current sensor to be diagnosed. The fault handling result of the current sensor to be diagnosed is determined by comparing the sampling data of the current sensor to be diagnosed with the state estimate value corresponding to one of the adjacent sensors. If the fault handling result of the current sensor to be diagnosed is normal, then the next level sensor of the current sensor to be diagnosed is determined as the new current sensor to be diagnosed according to the coupling relationship, and the process returns to the step of determining the fault handling result of the current sensor to be diagnosed by comparing the sampled data of the current sensor to be diagnosed with the state estimate value of one of the adjacent sensors, until the fault handling of all other sensors is completed.

3. The sensor fault handling method according to claim 2, characterized in that, The fault handling result of the current reference sensor is determined by comparing the sampled data of the current reference sensor with the state estimates of the adjacent sensors before and after it, including: The sampled data of the current reference sensor is compared with the state estimates of the adjacent sensors before and after it. If the difference between the sampled data of at least one current reference sensor and the state estimate of each of the adjacent sensors is less than the first threshold, then the fault handling result of the current reference sensor is determined to be normal. If the difference between the sampled data of the current reference sensor and the state estimate of each of the adjacent sensors is greater than or equal to the first threshold, then the fault handling result of the current reference sensor is determined to be a fault.

4. The sensor fault handling method according to claim 2, characterized in that, The fault handling result of the current sensor to be diagnosed is determined by comparing the sampled data of the current sensor to be diagnosed with the state estimate value of one of its adjacent sensors, including: If the difference between the sampled data of the current sensor to be diagnosed and the state estimate of one of its adjacent sensors is less than the second threshold, then the fault handling result of the current sensor to be diagnosed is determined to be normal. If the difference between the sampled data of the current sensor to be diagnosed and the state estimate of one of its adjacent sensors is greater than or equal to the second threshold, then the fault handling result of the current sensor to be diagnosed is determined to be a fault.

5. The sensor fault handling method according to claim 3, characterized in that, When the fault handling result of the current reference sensor is determined to be a fault, the method further includes: The first difference is obtained by comparing the sampled data of the current reference sensor with the state estimate of the previous adjacent sensor. The second difference is obtained by comparing the sampled data of the current reference sensor with the state estimate of the next adjacent sensor; If both the first difference and the second difference are less than the first threshold, then the fault handling results of the adjacent sensors are determined to be normal. Then, the state estimation values ​​corresponding to the adjacent sensors are used as the replacement sampling data of the current reference sensor for redundancy.

6. The sensor fault handling method according to claim 5, characterized in that, If either the first difference or the second difference is greater than or equal to the first threshold, the method further includes: If the difference between the state estimates of adjacent sensors is greater than the third threshold, the fault handling result of the adjacent sensors is determined to be a fault, and the system is shut down for protection.

7. The sensor fault handling method according to claim 5, characterized in that, The state estimates corresponding to each of the preceding and following adjacent sensors are used as replacement sampling data for the current reference sensor, including: A first weight parameter and a second weight parameter are preset for the state estimation values ​​corresponding to the adjacent sensors before and after; wherein, the second weight parameter is greater than the first weight parameter; The first sampling data is obtained by multiplying the first weight parameter with the state estimate corresponding to the previous adjacent sensor. The second weighting parameter is multiplied by the state estimate corresponding to the next adjacent sensor to obtain the second sampling data; The first sampled data and the second sampled data are summed to obtain the replacement sampled data of the current reference sensor.

8. The sensor fault handling method according to claim 4, characterized in that, When the fault handling result of the current sensor to be diagnosed is a fault, the method further includes: The state estimate of one of the adjacent sensors corresponding to the current sensor to be diagnosed is used as the replacement sampling data of the current sensor to be diagnosed for redundancy.

9. The sensor fault handling method according to claim 5 or 8, characterized in that, During the sensor redundancy processing, the method further includes: If another sensor of the target sensor that is redundantly processed fails, the auxiliary converter will be shut down for protection.

10. The sensor fault handling method according to claim 1, characterized in that, The process of obtaining the state estimate, the method further includes: Obtain historical sampling data; Feature parameters are obtained by performing time-series feature processing on the historical sampling data; Record the number of calls; call the artificial intelligence model, input the feature parameters into the artificial intelligence model, and output the current state estimate; when the current state estimate is not within the first preset range, increment the number of calls by 1, and return to the step of calling the artificial intelligence model for training, until the current state estimate is within the first preset range or the preset number of calls is reached before the number of calls is reached, thus completing the training process of the artificial intelligence model; Correspondingly, the current feature parameters are obtained by performing time-series feature processing on the sampled data, and the trained artificial model is called to input the current feature parameters into the trained artificial model to output the final state estimate.

11. A sensor fault handling device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the sensor fault handling method as described in any one of claims 1 to 10 when executing the computer program.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the sensor fault handling method as described in any one of claims 1 to 10.

Images