Fault detection method and system for charger

Abstract

The invention relates to the technical field of fault detection, and discloses a fault detection method and system for a charger, and the method comprises the steps: obtaining the charging state of a battery, and the charging state of the battery comprises the charging current, the charging power, the state of charge, and the temperature of the battery; performing feature extraction on the battery charging state, and constructing a charging feature matrix; inputting the charging characteristic matrix into a battery health prediction model to obtain a battery charging demand prediction value; performing state analysis and calculation according to the battery charging state to obtain a charger state; performing correlation analysis according to the charger state and the battery charging demand prediction value to obtain fault prediction information; and obtaining a charger fault detection result according to the fault prediction information. The method can improve the fault detection efficiency of the charger.

Description

Technical Field

[0001] The present invention relates to the technical field of fault detection, and in particular to a charger fault detection method and system. Background Art

[0002] At present, in recent years, the electric vehicle industry has developed rapidly. As an important part of electric vehicles, the technological progress of batteries and the development of charging facilities have attracted much attention. The current development status of battery technology is reflected in the rapid expansion of production capacity. It is expected that by 2025, the production capacity of power lithium batteries will exceed 3000GWh. In terms of charging technology, in order to meet the large demand for charging and improve charging efficiency and safety, the research and development of advanced charging technology and equipment has become crucial. The existing charger technology is developing in the direction of power level improvement, size miniaturization, two-way conversion and integration. For example, the output power of the on-board OBC (On-Board Charger) has been upgraded from 3.3kW to 6.6kW, and there is a trend of development towards 11kW and even 22kW. However, with the high power of the charger and the lack of personnel supervision, the failure rate of the charger is getting higher and higher. Due to the imperfection of technology and equipment, failures often occur during the charging process, causing safety problems, damaging batteries and equipment, and even causing harm to users.

[0003] In one prior art, the process of charger fault detection is carried out by direct observation and manual inspection, including a detailed inspection of the appearance of the charger to look for any obvious physical damage, detachment or abnormal phenomena, such as cable wear or loose plug contact. The diagnostic instrument is used to read the fault codes of the vehicle and the charger, which are stored in a database and list some common faults. Generally, the electrical parameters of the charger are measured using tools such as voltmeters and ammeters, and the measurement results are compared with the normal value range to determine whether there is an electrical fault. At the same time, communication fault detection is performed, and it is necessary to check whether the communication between the charger and the battery management system (BMS) is normal, because communication problems may cause the charging process to fail to proceed normally. Resistance detection involves measuring the resistance of the charging gun CC to the internal ground of the charging gun to ensure that it is within the normal value range. Waveform diagnosis uses a waveform data capture instrument to capture the CAN waveform and analyze the normality of the signal, such as whether the CAN-L signal is normal. Fault diagnosis message analysis involves interpreting the fault diagnosis messages exchanged between the charger and the BMS through the communication protocol. These messages contain fault codes, descriptions, levels and additional information. These cumbersome steps cause the problem of low fault detection efficiency in the prior art. Summary of the invention

[0004] The present invention provides a charger fault detection method, system, electronic equipment and storage medium to improve the charger fault detection efficiency.

[0005] In a first aspect, in order to solve the above technical problems, the present invention provides a charger fault detection method, comprising: Obtaining a battery charging state, wherein the battery charging state includes: charging current, charging power, state of charge, and battery temperature; Extracting features of the battery charging state and constructing a charging feature matrix; Inputting the charging characteristic matrix into a battery health prediction model to obtain a battery charging demand prediction value; Perform state analysis and calculation according to the battery charging state to obtain the charger state; Performing a correlation analysis between the charger state and the battery charging demand prediction value to obtain fault prediction information; A charger fault detection result is obtained according to the fault prediction information.

[0006] In an optional implementation, extracting features of the battery charging state and constructing a charging feature matrix includes: Normalizing the battery charging state to obtain a normalized charging current and a normalized charging power; Calculate the normalized current difference between two adjacent time points and add them up to get the average current increasing sequence, including:

[0007] in, is the average current increasing sequence, k Number the time points. n is the total number of time points, for k Normalized charging current at time point 1; Calculate the power difference between two adjacent time points and take the average value to get the average power series, including:

[0008] in, is the average power sequence, for k Normalized charging power at time point 1; The state of charge, the battery temperature, the average current increasing sequence and the average power sequence are combined into a charging characteristic matrix.

[0009] In an optional implementation manner, inputting the charging characteristic matrix into a battery health prediction model to obtain a battery charging demand prediction value includes: A battery health prediction model is constructed based on historical charging data, and the model is trained. The training is determined to be complete when the loss function value of the detection model is less than a preset loss threshold or the current training times reaches a preset training times, thereby obtaining a trained model. Inputting the charging feature matrix into the trained model to obtain a battery charging demand prediction value; The battery charging demand prediction value includes the battery health status, remaining service life and battery status prediction value; The predicted state of charge in the battery state prediction value is calculated by the following formula: ;in, express k +1 time state of charge prediction value, express k The actual value of the state of charge at the moment, C Indicates the rated capacity of the battery. n Indicates the total number of time points, k Indicates the time point number. express k The instantaneous current at time, a , c is the model coefficient, is the instantaneous current of the integration process, for k+ The instantaneous current at time 1, t is the integral process quantity.

[0010] In an optional implementation, performing state analysis and calculation according to the battery charging state to obtain the charger state includes: The charger state includes charger current and charger power; The charger current is calculated using the following formula:

[0011] in, is the charger current, is the total number of time points, k Number the time points. for k No. time charger current, The current consumed by the charger; The charger power is calculated using the following formula:

[0012] in, is the charger power, for k No. time charger power, is the power loss coefficient of the charger.

[0013] In an optional implementation manner, performing a correlation analysis based on the charger state and the battery charging demand prediction value to obtain fault prediction information includes: The fault prediction information is calculated by the following formula:

[0014] in, is the fault prediction information, is the weight of the correlation between the battery charging demand prediction value and the charger current in the corresponding charger state, is the weight of the correlation between the battery charging demand prediction value and the charger power in the corresponding charger state, is the battery charging demand forecast value, is the charger current, is the charger power, is the correlation between the battery charging demand prediction value and the charger current in the corresponding charger state, The battery charging demand prediction value and the charger power correlation in the corresponding charger state.

[0015] In an optional implementation manner, obtaining a charger fault detection result according to the fault prediction information includes: When the fault prediction information is a negative number, it indicates that the charger has a fault; When the fault prediction information is a positive number, it indicates that the charger is in normal use.

[0016] In an optional implementation, the method for calculating the charger loss current includes: The total charging time is calculated using the following formula:

[0017] in, is the total charging time, n is the total number of time points, k Number the time points. for k Charging time of the number time; The charger loss current is calculated using the following formula:

[0018] in, is the charger consumption current, is the total power of the charger, is the total boot time.

[0019] In a second aspect, the present invention provides a fault detection system for a charger, comprising: A data acquisition module, used to acquire a battery charging state, wherein the battery charging state includes: charging current, charging power, state of charge, and battery temperature; A feature extraction module, used to extract features of the battery charging state and construct a charging feature matrix; A state prediction module, used for inputting the charging characteristic matrix into a battery health prediction model to obtain a battery charging demand prediction value; A charger analysis module, used to perform status analysis calculation according to the battery charging status to obtain a charger status; A fault information module, used to perform a correlation analysis based on the charger state and the battery charging demand prediction value to obtain fault prediction information; The result output module is used to obtain the charger fault detection result according to the fault prediction information.

[0020] In a third aspect, the present invention further provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements any one of the above-mentioned charger fault detection methods when executing the computer program.

[0021] In a fourth aspect, the present invention further provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned charger fault detection methods.

[0022] Compared with the prior art, the present invention has the following beneficial effects: the present invention discloses a charger fault detection method, the method comprising: obtaining a battery charging state, the battery charging state comprising: charging current, charging power, charge state, battery temperature; performing feature extraction on the battery charging state to construct a charging feature matrix; inputting the charging feature matrix into a battery health prediction model to obtain a battery charging demand prediction value; performing state analysis and calculation according to the battery charging state to obtain a charger state; performing correlation analysis between the charger state and the battery charging demand prediction value to obtain fault prediction information; and obtaining a charger fault detection result according to the fault prediction information.

[0023] This method first obtains the battery charging state, including key parameters such as charging current, charging power, state of charge (SOC), battery temperature, etc., and extracts features of these battery charging states to construct a charging feature matrix. This process can reduce data redundancy and improve model processing efficiency. At the same time, through feature selection and engineering processing, it can better capture the key factors that have a significant impact on the battery health state. Subsequently, the constructed charging feature matrix is ​​input into the pre-trained battery health prediction model to obtain the battery charging demand prediction value, including the battery health state (SOH) and remaining service life (RUL). This step relies on the prediction model and can evaluate the current condition and future trend of the battery. Further, according to the obtained battery charging state, the charger analysis method is used to calculate the charger state. This process can obtain the output electrical parameters of the charger for subsequent fault detection. Combined with the charger state and the battery charging demand prediction value, the correlation analysis method is used to obtain fault prediction information. This step quantifies the correlation between the battery charging demand prediction value and the motor state. Compared with the traditional threshold detection method, this method can detect potential problems earlier and reduce false alarms and missed alarms. At the same time, by learning from a large amount of data, the weights can be continuously optimized to improve the accuracy of the fault prediction model.

[0024] Specifically, in the process of calculating the charging feature matrix, the normalization process can eliminate the influence of different dimensions and scales, so that different types of charging state data are in the same order of magnitude, which is convenient for subsequent data analysis and modeling. It can improve the training efficiency of the model and enhance the generalization ability of the model. By calculating the current difference and power difference at adjacent time points, the changing trends of current and power during the charging process can be captured. These changing trends directly reflect the working state of the battery and can reduce errors. In the process of calculating the average current increasing sequence and the average power sequence, the noise in the original data can be effectively smoothed through accumulation and averaging operations, the influence of accidental factors can be reduced, and the features can be made more reliable. The state of charge, battery temperature, average current increasing sequence and average power sequence are combined into a charging feature matrix to realize the fusion of multi-dimensional features. This multi-dimensional feature matrix not only contains the basic working state of the battery, but also integrates dynamic change information, which can more comprehensively describe the operating status of the battery and more accurately predict the subsequent battery state.

[0025] In summary, this method predicts the battery working state through the battery health prediction model and combines the correlation analysis algorithm of the charger state to improve the fault detection efficiency and accuracy of the charger. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a schematic flow chart of a charger fault detection method provided by the first embodiment of the present invention; Figure 2It is a schematic diagram of the structure of a fault detection system for a charger provided in a second embodiment of the present invention. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] At present, in recent years, the electric vehicle industry has developed rapidly. As an important component of electric vehicles, the technological progress of batteries and the development of charging facilities have attracted much attention. The current development status of battery technology is reflected in the rapid expansion of production capacity. It is estimated that by 2025, my country's power lithium battery production capacity may exceed 3000GWh. In terms of charging technology, in order to meet the large demand for charging and improve charging efficiency and safety, the research and development of advanced charging technology and equipment has become crucial. The existing charger technology is developing in the direction of power level improvement, size miniaturization, two-way conversion and integration. For example, the output power of the on-board OBC (On-Board Charger) has been upgraded from 3.3kW to 6.6kW, and there is a trend towards 11kW and even 22kW. However, with the high power of the charger and the lack of personnel supervision, the failure rate of the charger is getting higher and higher. Due to the imperfection of technology and equipment, failures often occur during the charging process, causing safety problems, damaging batteries and equipment, and even causing harm to users.

[0029] In one prior art, the process of charger fault detection is carried out by direct observation and manual inspection, including a detailed inspection of the appearance of the charger to look for any obvious physical damage, detachment or abnormal phenomena, such as cable wear or loose plug contact. The diagnostic instrument is used to read the fault codes of the vehicle and the charger, which are stored in a database and list some common faults. Generally, the electrical parameters of the charger are measured using tools such as voltmeters and ammeters, and the measurement results are compared with the normal value range to determine whether there is an electrical fault. At the same time, communication fault detection is performed, and it is necessary to check whether the communication between the charger and the battery management system (BMS) is normal, because communication problems may cause the charging process to fail to proceed normally. Resistance detection involves measuring the resistance of the charging gun CC to the internal ground of the charging gun to ensure that it is within the normal value range. Waveform diagnosis uses a waveform data capture instrument to capture the CAN waveform and analyze the normality of the signal, such as whether the CAN-L signal is normal. Fault diagnosis message analysis involves interpreting the fault diagnosis messages exchanged between the charger and the BMS through the communication protocol. These messages contain fault codes, descriptions, levels and additional information. These cumbersome steps cause the problem of low fault detection efficiency in the prior art.

[0030] To solve the above problems, refer to Figure 1 The first embodiment of the present invention provides a method for detecting a fault of a charger, comprising the following steps: S11, obtaining a battery charging state, wherein the battery charging state includes: charging current, charging power, state of charge, and battery temperature; S12, extracting features of the battery charging state and constructing a charging feature matrix; S13, inputting the charging characteristic matrix into a battery health prediction model to obtain a battery charging demand prediction value; S14, performing state analysis and calculation according to the battery charging state to obtain a charger state; S15, performing a correlation analysis based on the charger state and the battery charging demand prediction value to obtain fault prediction information; S16, obtaining a charger fault detection result according to the fault prediction information.

[0031] In step S11, the battery charging state is acquired, and the battery charging state includes: charging current, charging power, state of charge, and battery temperature.

[0032] In one embodiment, obtaining the battery charging state involves the use of a variety of sensors and monitoring devices. Specifically, the charging current and charging power can be measured by high-precision current transformers and power meters, which can monitor and record the current and power changes during the charging process in real time. The state of charge (SOC) is calculated by the battery management system's ampere-hour integration method and open circuit voltage method, which is calculated based on parameters such as battery voltage, current and temperature. The battery temperature is measured by thermistors or thermocouple sensors installed inside or on the surface of the battery pack, which provide accurate temperature readings. The detected data is transmitted to the central control system or data acquisition system in the form of digital signals, which will perform preliminary processing and storage on the data for subsequent analysis. This method performs fault detection by directly calling the data stored in the system.

[0033] It is worth noting that the battery charging status data is indexed by timestamps, and each time point corresponds to multiple sets of measurement values, including: charging current, charging power, state of charge, and battery temperature. It is stored in the form of a time series database, which is specially designed to store and query time series data for real-time monitoring and analysis of battery status.

[0034] In step S12, the battery charging state is subjected to feature extraction to construct a charging feature matrix.

[0035] In one implementation, the battery charging state is normalized to obtain a normalized charging current and a normalized charging power; Calculate the normalized current difference between two adjacent time points and add them up to get the average current increasing sequence, including:

[0036] in, is the average current increasing sequence, k Number the time points. n is the total number of time points, for k Normalized charging current at time point 1; Calculate the power difference between two adjacent time points and take the average value to get the average power series, including:

[0037] in, is the average power sequence, for k Normalized charging power at time point 1; The state of charge, the battery temperature, the average current increasing sequence and the average power sequence are combined into a charging characteristic matrix.

[0038] It is worth noting that the normalization process is to eliminate the influence of different dimensions and scales so that different types of charging state data are in the same order of magnitude. It can avoid some features dominating the model training process because of their large values, and ensure that the contribution of each feature in the model is more balanced. This method normalizes the current value by dividing it by its maximum current value and the power by its maximum power value. In order to avoid the influence of noise on the normalization process, before the normalization process, this method removes excessive noise by smoothing and denoising to avoid the influence of noise. Furthermore, the average current increasing sequence can reflect the cumulative change of current and provide dynamic information about the battery charging behavior. By calculating and accumulating the normalized current difference at adjacent time points, the dynamic change trend of current during the charging process can be captured. This helps to identify sudden changes and anomalies in current, which are often early signs of battery failure. By calculating the power difference at adjacent time points and taking the average value, the change trend of power during the charging process can be captured. The change in power can also reflect the working state and potential problems of the battery. At the same time, the average power sequence provides a smooth estimate of power changes, which helps to filter out short-term noise and fluctuations and enhance the stability of the data. Furthermore, the time series data is converted to the frequency domain through Fourier transform or wavelet transform to extract frequency domain features. The periodicity and oscillation characteristics in the data can be captured, which helps to identify specific failure modes. Calculating the statistical characteristics of time series data can provide overall information about the data distribution and help identify data anomalies and changing trends.

[0039] In step S13, the charging characteristic matrix is ​​input into a battery health prediction model to obtain a battery charging demand prediction value.

[0040] In one implementation, a battery health prediction model is constructed based on historical charging data, the model is trained, and the training is determined to be complete after the loss function value of the detection model is less than a preset loss threshold or the current training number reaches a preset training number, thereby obtaining a trained model; Inputting the charging feature matrix into the trained model to obtain a battery charging demand prediction value; The battery charging demand prediction value includes the battery health status, remaining service life and battery status prediction value; The predicted state of charge in the battery state prediction value is calculated by the following formula: ; in, express k +1 time state of charge prediction value, express k The actual value of the state of charge at the moment, C Indicates the rated capacity of the battery. n Indicates the total number of time points,k Indicates the time point number. express k The instantaneous current at time, a , c is the model coefficient, is the instantaneous current of the integration process, for k+ The instantaneous current at time 1, t is the integral process quantity.

[0041] It is worth noting that the predicted state of charge formula combines the actual state of charge, the rated capacity of the battery, the instantaneous current and the model coefficient to predict the state of charge at the next moment. The integral term in the formula represents the time from the point in time n To the time point n+ 1, which reflects the change in the net charge of the battery during this period. This integral can be approximated by numerical methods. The logic of the entire formula is: when the current state of charge and instantaneous current are known, the state of charge at the next moment is predicted by adding the current integral term (representing the change in the net charge of the battery during this period) and adjusting the model coefficient. The model coefficient is obtained by training with historical charging data.

[0042] In one embodiment, a long short-term memory network (LSTM) model is used as a battery health prediction model, and historical charging data includes charging current, charging power, state of charge (SOC), battery temperature, etc. The data set is divided into a training set, a validation set, and a test set, with a ratio of 70%, 15%, and 15%. This method is not limited to this. The optimizer selects 'adam', and the loss function selects the mean square error loss. Set the number of training times, and use EarlyStopping as a callback function to monitor the training process. Through the above steps, a battery health prediction model can be built and trained based on historical charging data, and finally a model that can predict the battery state of health (SOH) and remaining service life (RUL) is obtained. The whole process includes data preparation, model selection, model training, model evaluation, and model prediction. During the model training process, GridSearchCV is used to search for hyperparameters to find the best hyperparameter combination. Add L1 or L2 regularization terms to the loss function to prevent overfitting. Furthermore, the learning rate decay strategy is used to gradually reduce the learning rate to improve the training effect.

[0043] S14, performing state analysis and calculation according to the battery charging state to obtain a charger state.

[0044] In one implementation, the charger state includes charger current and charger power; The charger current is calculated using the following formula:

[0045] in, is the charger current, is the total number of time points, k Number the time points. for k No. time charger current, The current consumed by the charger; The charger power is calculated using the following formula:

[0046] in, is the charger power, is the total number of time points, k Number the time points. for k No. time charger power, is the power loss coefficient of the charger.

[0047] The total charging time is calculated using the following formula:

[0048] in, is the total charging time, is the total number of time points, k Number the time points. for k Charging time of the number time; The charger loss current is calculated using the following formula:

[0049] in, is the charger consumption current, is the total power of the charger, is the total boot time, The total charging time.

[0050] It is worth noting that the power loss coefficient of the charger is a constant that reflects the loss of the charger during the power transmission process. It is obtained by measuring the ratio of the charger output power to the actual charging power of the battery. The charger will have a certain loss during the power transmission process, and this loss coefficient is used to adjust the average power value to reflect the actual charger power output. The loss coefficient is a value less than 1, indicating that the actual output power is lower than the theoretical input power. The total power of the charger is the total power consumed by the charger during the entire charging process. It can be measured by a power meter or an energy meter. The total startup time is the total time from the start to the stop of the charger, which is calculated by the system time. The average value of the charging current at all time points can reflect the overall level of the current during the charging process. As the charger current value, the error can be reduced. Furthermore, the charger will have a certain amount of energy loss during the operation process, and this part of the loss will be converted into additional current. Therefore, when calculating the charger current, this loss current needs to be added. The charger loss current is obtained by dividing the total power of the charger by the non-charging time and taking the square root. This formula shows that the loss current is proportional to the power loss during the non-charging time.

[0051] In step S15, a correlation analysis is performed based on the charger state and the battery charging demand prediction value to obtain fault prediction information.

[0052] In one implementation, the fault prediction information is calculated using the following formula: ; in, is the fault prediction information, is the weight of the correlation between the battery charging demand prediction value and the charger current in the corresponding charger state, is the weight of the correlation between the battery charging demand prediction value and the charger power in the corresponding charger state, is the battery charging demand forecast value, is the charger current, is the charger power, is the correlation between the battery charging demand prediction value and the charger current in the corresponding charger state, The battery charging demand prediction value and the charger power correlation in the corresponding charger state.

[0053] It is worth noting that the fault prediction information is the final calculation result. It is a comprehensive indicator used to predict faults that occur during battery charging. The weight coefficient is used to measure the importance of the correlation between the predicted value of battery charging demand and the charger current in fault prediction. It is determined based on actual data and model training results to reflect the contribution of the charger current to fault prediction. This is a statistical function used to calculate the correlation coefficient between two variables. In this formula, the Corr function is used to calculate the correlation between the predicted value of battery charging demand and the charger current and power. The range of the correlation coefficient is [-1, 1], where 1 indicates a complete positive correlation, -1 indicates a complete negative correlation, and 0 indicates no linear correlation. It is the predicted value of battery charging demand, which is output by the above battery health prediction model.

[0054] In one implementation, the Pearson correlation is used as the correlation function and is calculated using the following formula:

[0055] in, is the correlation function, i is the sample number, n is the total number of samples, for i The first sample of for i The second sample number, is the first sample mean, is the second sample mean. In this method, for , for or .

[0056] S16, obtaining a charger fault detection result according to the fault prediction information.

[0057] In one implementation, when the fault prediction information is a negative number, it indicates that there is a fault in the charger; and when the fault prediction information is a positive number, it indicates that the charger is in normal use.

[0058] It is worth noting that when the fault prediction information is a positive number, this indicates that there is a positive correlation between the predicted value of the battery charging demand and the charger current and charger power. When the battery demand increases, the charger current and power also increase accordingly, which is in line with the normal working mode, indicating that the charger is working normally. When there is a negative correlation between the two, this indicates that when the battery demand increases, the charger current and power decrease instead, or when the battery demand decreases, the charger current and power increase. This abnormal pattern indicates that the charger is faulty and cannot respond normally to the battery demand. The role of the weight coefficient is to adjust the contribution of current and power. When the change of the charger current is more important for fault detection, the proportion of the weight coefficient one can be increased.

[0059] In one implementation, when the charger fails, an alarm may be triggered to notify maintenance personnel to perform inspection and repair.

[0060] In order to facilitate the understanding of the present invention, some preferred embodiments of the present invention are further described below.

[0061] In one implementation, a fault tree analysis method can be used to identify faults. Define a top event, which indicates the most undesirable fault event, such as "the charger cannot charge normally". Then find all possible direct causes of the top event, and continue to decompose the intermediate events until all basic causes are found. Further, use logic gates (such as AND gates and OR gates) to connect the top event, the intermediate event, and the bottom event to form a fault tree. Through fault tree analysis, the minimum cut set that causes the top event to occur is determined, and the critical fault path is found. Calculate the probability of the top event and evaluate the importance of each bottom event. For example: top event: the charger cannot charge normally. Intermediate events: power supply problem, communication problem, internal circuit failure, sensor failure. Bottom events: unstable power supply voltage, power failure, communication interruption, communication protocol error, control circuit failure, power device failure, temperature sensor failure, current sensor failure. A fault tree can be constructed. Through fault tree analysis, the minimum cut set that causes the top event can be found, such as unstable power supply voltage AND communication interruption and control circuit failure OR power device failure. The probability of the bottom layer is calculated again, and the probability of unstable power supply voltage is 0.05. The probability of the remaining events is not explained, and the present invention does not limit it. The probability of the top event is calculated by the fault tree, for example, 0.26. The common fault modes of the charger can be systematically identified by fault tree analysis (FTA).

[0062] In summary, the present invention proposes a charger fault detection method and system, aiming to improve the efficiency and accuracy of charger fault detection. The method obtains the battery charging state, including key parameters such as charging current, charging power, state of charge and battery temperature, and extracts features of these states to construct a charging feature matrix. This process not only reduces data redundancy and improves model processing efficiency, but also better captures key factors that have a significant impact on the battery health state through feature selection and engineering processing. Subsequently, the constructed charging feature matrix is ​​input into the pre-trained battery health prediction model to obtain the battery charging demand prediction value, including the battery health state (SOH) and remaining service life (RUL). This step relies on the prediction model and can evaluate the current condition and future trend of the battery. Furthermore, according to the obtained battery charging state, the charger analysis method is used to calculate the charger state. This process can obtain the output electrical parameters of the charger for subsequent fault detection. Combined with the charger state and the battery charging demand prediction value, the fault prediction information is obtained by using the correlation analysis method. This step quantifies the correlation between the battery charging demand prediction value and the motor state. Compared with the traditional threshold detection method, this method can detect potential problems earlier and reduce false positives and false negatives. At the same time, by learning a large amount of data, the weights can be continuously optimized to improve the accuracy of the fault prediction model. In the process of calculating the charging feature matrix, the normalization process eliminates the influence of different dimensions and scales, so that different types of charging state data are in the same order of magnitude, which is convenient for subsequent data analysis and modeling. By calculating the current difference and power difference at adjacent time points, the changing trends of current and power during the charging process can be captured. These changing trends directly reflect the working state of the battery and can reduce errors. In the process of calculating the average current increasing sequence and the average power sequence, the noise in the original data can be effectively smoothed by accumulation and averaging operations, reducing the influence of accidental factors and making the features more reliable. The state of charge, battery temperature, average current increment sequence and average power sequence are combined into a charging feature matrix to achieve the fusion of multi-dimensional features. This multi-dimensional feature matrix not only contains the basic working state of the battery, but also integrates dynamic change information, which can more comprehensively describe the operating status of the battery and more accurately predict the subsequent battery status. The battery health prediction model in this scheme can be based on machine learning models such as long short-term memory network (LSTM) and trained with historical charging data to predict the health state and remaining service life of the battery. The calculation of the charger status involves the analysis of the charger current and power. By calculating the total charging time and the charger loss current, the actual working state of the charger can be obtained, which provides a basis for fault detection. Finally, according to the fault prediction information, the fault detection result of the charger can be obtained. When the fault prediction information is negative, it means that the charger has a fault; when it is positive, it means that the charger is in normal use.This data-driven approach can provide more accurate and timely fault detection by real-time monitoring and analysis of battery and charger status, thereby improving the reliability and safety of the charger.

[0063] In summary, this scheme builds a charger fault detection system by comprehensively using technologies such as battery status monitoring, feature extraction, machine learning prediction, and correlation analysis. This method can improve the efficiency of charger fault detection.

[0064] Reference Figure 2 A second embodiment of the present invention provides a fault detection system for a charger, comprising: A data acquisition module, used to acquire a battery charging state, wherein the battery charging state includes: charging current, charging power, state of charge, and battery temperature; A feature extraction module, used to extract features of the battery charging state and construct a charging feature matrix; A state prediction module, used for inputting the charging characteristic matrix into a battery health prediction model to obtain a battery charging demand prediction value; A charger analysis module, used to perform status analysis calculation according to the battery charging status to obtain a charger status; A fault information module, used to perform a correlation analysis based on the charger state and the battery charging demand prediction value to obtain fault prediction information; The result output module is used to obtain the charger fault detection result according to the fault prediction information.

[0065] Preferably, the data acquisition module is used to: Obtain the battery charging status, which includes: charging current, charging power, state of charge, and battery temperature.

[0066] Preferably, the feature extraction module is used to: Extracting features of the battery charging state and constructing a charging feature matrix include: Normalizing the battery charging state to obtain a normalized charging current and a normalized charging power; Calculate the normalized current difference between two adjacent time points and add them up to get the average current increasing sequence, including:

[0067] in, is the average current increasing sequence, k Number the time points. n is the total number of time points, for k Normalized charging current at time point 1; Calculate the power difference between two adjacent time points and take the average value to get the average power series, including:

[0068] in, is the average power sequence, k Number the time points. n is the total number of time points, for k Normalized charging power at time point 1; The state of charge, the battery temperature, the average current increasing sequence and the average power sequence are combined into a charging characteristic matrix.

[0069] Preferably, the state prediction module is used to: The charging characteristic matrix is ​​input into the battery health prediction model to obtain the battery charging demand prediction value, including: A battery health prediction model is constructed based on historical charging data, and the model is trained. The training is determined to be complete when the loss function value of the detection model is less than a preset loss threshold or the current training times reaches a preset training times, thereby obtaining a trained model. Inputting the charging feature matrix into the trained model to obtain a battery charging demand prediction value; The battery charging demand prediction value includes the battery health status, remaining service life and battery status prediction value; The predicted state of charge in the battery state prediction value is calculated by the following formula: ; in, express k +1 time state of charge prediction value, express k The actual value of the state of charge at the moment, C Indicates the rated capacity of the battery. n Indicates the total number of time points, k Indicates the time point number. express k The instantaneous current at time, a , c is the model coefficient. is the instantaneous current of the integration process, for k+ The instantaneous current at time 1, t is the integral process quantity.

[0070] Preferably, the charger analysis module is used to: Performing state analysis and calculation according to the battery charging state to obtain the charger state includes: The charger state includes charger current and charger power; The total charging time is calculated using the following formula:

[0071] in, is the total charging time, is the total number of time points, k is the time point number, is the charging time at time k; The charger loss current is calculated using the following formula:

[0072] in, is the charger consumption current, is the total power of the charger, is the total boot time, Is the total charging time.

[0073] The charger current is calculated using the following formula:

[0074] in, is the charger current, is the total number of time points, k Number the time points. for k No. time charger current, The current consumed by the charger; The charger power is calculated using the following formula:

[0075] in, is the charger power, for k No. time charger power, is the power loss coefficient of the charger.

[0076] Preferably, the fault information module is used to perform a correlation analysis based on the charger state and the battery charging demand prediction value to obtain fault prediction information, including: The fault prediction information is calculated by the following formula: ; in, is the fault prediction information, is the weight of the correlation between the battery charging demand prediction value and the charger current in the corresponding charger state, is the weight of the correlation between the battery charging demand prediction value and the charger power in the corresponding charger state, is the battery charging demand forecast value, is the charger current, is the charger power, is the correlation between the battery charging demand prediction value and the charger current in the corresponding charger state, The battery charging demand prediction value and the charger power correlation in the corresponding charger state.

[0077] Preferably, the result output module is used to obtain the charger fault detection result according to the fault prediction information, including: When the fault prediction information is a negative number, it indicates that the charger has a fault; When the fault prediction information is a positive number, it indicates that the charger is in normal use.

[0078] It should be noted that a charger fault detection system provided in an embodiment of the present invention is used to execute all process steps of a charger fault detection method in the above embodiment, and the working principles and beneficial effects of the two correspond one to one, so they will not be repeated here.

[0079] The embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a data acquisition program. When the processor executes the computer program, the steps in the above-mentioned embodiments of the charger fault detection method are implemented, such as Figure 1 Alternatively, when the processor executes the computer program, the functions of the modules / units in the above-mentioned device embodiments are realized, such as the data acquisition module.

[0080] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the electronic device.

[0081] The electronic device may be a computing device such as a desktop computer, a notebook, a PDA, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. The electronic device may include more or fewer components than the above components, or may combine certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0082] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, and uses various interfaces and lines to connect various parts of the entire electronic device.

[0083] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the electronic device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0084] Wherein, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0085] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0086] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A charger fault detection method, characterized in that: include: Obtaining a battery charging state, wherein the battery charging state includes: charging current, charging power, state of charge, and battery temperature; Extracting features of the battery charging state and constructing a charging feature matrix; Inputting the charging characteristic matrix into a battery health prediction model to obtain a battery charging demand prediction value; Perform state analysis and calculation according to the battery charging state to obtain the charger state; Performing a correlation analysis between the charger state and the battery charging demand prediction value to obtain fault prediction information; A charger fault detection result is obtained according to the fault prediction information.

2. The charger fault detection method according to claim 1, characterized in that: The extracting features of the battery charging state and constructing a charging feature matrix includes: Normalizing the battery charging state to obtain a normalized charging current and a normalized charging power; Calculate the normalized current difference between two adjacent time points and add them up to get the average current increasing sequence, including: ;in, is the average current increasing sequence, k Number the time points. n is the total number of time points, for k Normalized charging current at time point 1; Calculate the power difference between two adjacent time points and take the average value to get the average power series, including: ;in, is the average power sequence, for k Normalized charging power at time point 1; The state of charge, the battery temperature, the average current increasing sequence and the average power sequence are combined into a charging characteristic matrix.

3. The charger fault detection method according to claim 1, characterized in that: The step of inputting the charging characteristic matrix into a battery health prediction model to obtain a battery charging demand prediction value includes: A battery health prediction model is constructed based on historical charging data, and the model is trained. The training is determined to be complete when the loss function value of the detection model is less than a preset loss threshold or the current training times reaches a preset training times, thereby obtaining a trained model. Inputting the charging feature matrix into the trained model to obtain a battery charging demand prediction value; The battery charging demand prediction value includes the battery health status, remaining service life and battery status prediction value; The predicted state of charge in the battery state prediction value is calculated by the following formula: ;in, express k +1 time state of charge prediction value, express k The actual value of the state of charge at the moment, C Indicates the rated capacity of the battery. n Indicates the total number of time points, k Indicates the time point number. express k The instantaneous current at time a , c is the model coefficient is the instantaneous current of the integration process, for k+ The instantaneous current at time 1, t is the integral process quantity.

4. The charger fault detection method according to claim 1, characterized in that: The performing state analysis and calculation according to the battery charging state to obtain the charger state includes: The charger state includes charger current and charger power; The charger current is calculated using the following formula: ;in, is the charger current, is the total number of time points, k Number the time points. for k Number time charger current, The current consumed by the charger; The charger power is calculated using the following formula: ;in, is the charger power, for k No. time charger power, is the power loss coefficient of the charger.

5. The charger fault detection method according to claim 1, characterized in that: The performing correlation analysis on the charger state and the battery charging demand prediction value to obtain fault prediction information includes: The fault prediction information is calculated by the following formula: ;in, is the fault prediction information, is the weight of the correlation between the battery charging demand prediction value and the charger current in the corresponding charger state, is the weight of the correlation between the battery charging demand prediction value and the charger power in the corresponding charger state, is the battery charging demand forecast value, is the charger current, The charger power is the correlation between the battery charging demand prediction value and the charger current in the corresponding charger state, The battery charging demand prediction value and the charger power correlation in the corresponding charger state.

6. The charger fault detection method according to claim 1, characterized in that: The step of obtaining a charger fault detection result according to the fault prediction information includes: When the fault prediction information is a negative number, it indicates that the charger has a fault; When the fault prediction information is a positive number, it indicates that the charger is in normal use.

7. The charger fault detection method according to claim 4, characterized in that: The calculation method of the charger loss current includes: The total charging time is calculated using the following formula: ;in, is the total charging time, is the total number of time points, k Number the time points. for k Charging time of the number time; The charger loss current is calculated using the following formula: ;in, is the charger consumption current, is the total power of the charger, is the total boot time.

8. A charger fault detection system, characterized in that: include: A data acquisition module, used to acquire a battery charging state, wherein the battery charging state includes: charging current, charging power, state of charge, and battery temperature; A feature extraction module, used to extract features of the battery charging state and construct a charging feature matrix; A state prediction module, used for inputting the charging characteristic matrix into a battery health prediction model to obtain a battery charging demand prediction value; A charger analysis module, used to perform status analysis calculation according to the battery charging status to obtain a charger status; A fault information module, used to perform a correlation analysis based on the charger state and the battery charging demand prediction value to obtain fault prediction information; The result output module is used to obtain the charger fault detection result according to the fault prediction information.

9. An electronic device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the charger fault detection method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the charger fault detection method according to any one of claims 1 to 7.

Images