Battery safety management method and system based on BMS

Abstract

The invention relates to the technical field of chargers, in particular to a battery safety management method and system based on a BMS. The method comprises the following steps: acquiring battery operation data in real time, segmenting continuous data flow by utilizing a sliding window technology, extracting a discretized state point set, extracting features from the discrete state point set and performing denoising processing to obtain an optimized feature vector set, performing mode division by adopting a clustering algorithm, and matching an error mode library to judge a potential error type. The method comprises the following steps: determining a corresponding relation between an abnormal cause and an operation stage by combining a decision tree analysis time and state dimension relevance, calculating and evaluating error severity through conditional probability, identifying an error stage, finally extracting a high-risk state point, comparing historical data to verify accuracy, and outputting a safety management analysis report. According to the invention, comprehensive monitoring and abnormal early warning of the operation state of the battery are realized, the use safety and reliability of the battery are effectively improved, and an important basis is provided for optimization of a battery management system.

Description

Technical Field

[0001] The present invention belongs to the technical field of battery management and safety control, and specifically relates to a battery safety management method and system based on BMS. Background Art

[0002] When dealing with continuous data streams, traditional battery management systems rely on simple threshold judgments for anomaly recognition. In the face of complex error patterns, their response capabilities are limited. Especially during the operation of the battery, the dynamic and multi-causal nature of error states makes it easy for traditional methods to mix anomalies with different causes, which may lead to misjudgments or missed judgments and affect the effect of refined management. In addition, existing technologies have insufficient accuracy in feature extraction and error pattern matching, and anomalies with the same appearance may stem from different operating stages or conditions. If their causes cannot be effectively distinguished, potential risks may be amplified. Therefore, how to construct a refined error analysis framework through time segmentation and state division has become an important direction for improving the safety management ability of BMS. Summary of the Invention

[0003] To solve the above technical problems, this application is proposed. Embodiments of the present invention provide a battery safety management method and system based on BMS. First, it collects voltage, current, and temperature parameters during the operation of the battery in real time through sensors to form a continuous data stream, and uses a time series sampling method to obtain an original data set; then, it uses the sliding window technique to segment the continuous data stream, extracts subsequences according to a preset time interval in the time dimension to obtain a discretized state point set; then, it extracts statistical features from the discretized state point set, including mean, variance, and change rate, and generates an initial feature vector set in combination with a physical model; subsequently, if there is noise interference in the initial feature vector set, the feature vectors are denoised through wavelet transform to obtain an optimized feature vector set; then, for the optimized feature vector set, the K-means clustering algorithm is used for pattern division, and the potential error type distribution is judged by matching the pre-established error pattern library according to the clustering results; then, the decision tree classification method is used to analyze the correlation between the time dimension and the state dimension to determine the corresponding relationship between the anomaly cause and the operating stage; finally, the conditional probability calculation method is used to evaluate the error severity of each state point to obtain the error stage recognition result, and the recognition accuracy is verified by comparing with historical data, and a final safety management analysis report is output. In this way, the error detection accuracy and safety management ability of BMS in complex dynamic scenarios can be effectively improved.

[0004] According to one aspect of the present invention, there is provided a battery safety management method based on BMS, which includes:

[0005] Obtain the battery operation data collected in real time by the sensor, including voltage, current and temperature parameters, to form a continuous data stream, and use the time series sampling method to obtain the original data set;

[0006] Use the sliding window technique to segment the continuous data stream, extract subsequences according to the preset time interval in the time dimension, and obtain a discretized state point set;

[0007] Extract statistical features from the discretized state point set, including mean, variance and change rate, and generate an initial feature vector set in combination with the physical model;

[0008] If there is noise interference in the initial feature vector set, denoise the feature vectors by the wavelet transform method to obtain an optimized feature vector set;

[0009] For the optimized feature vector set, use the K-means clustering algorithm for pattern division, and match the pre-established error pattern library according to the clustering results to judge the distribution of potential error types;

[0010] Analyze the correlation between the time dimension and the state dimension through the decision tree classification method to determine the corresponding relationship between the abnormal cause and the operation stage;

[0011] Use the conditional probability calculation method to evaluate the error severity of each state point to obtain the error stage recognition result;

[0012] Extract high-risk state points from the error stage recognition result, verify the recognition accuracy through comparison with historical data, and output the final safety management analysis report.

[0013] In the above battery safety management method based on BMS, using the sliding window technique to segment the continuous data stream, extracting subsequences according to the preset time interval in the time dimension to obtain a discretized state point set, including:

[0014] Set the step size and window width of the sliding window, and divide the continuous data stream into multiple overlapping or non-overlapping time periods;

[0015] Extract subsequences in each time period, and map the subsequences to a multi-dimensional feature space to form a discretized state point set;

[0016] Among them, the step size and window width of the sliding window are dynamically adjusted according to the battery operation characteristics and data sampling frequency.

[0017] In the above battery safety management method based on BMS, extracting statistical features from the discretized state point set, including mean, variance and change rate, and generating an initial feature vector set in combination with the physical model, including:

[0018] Calculate the mean and variance of each subsequence as the basic statistical features characterizing the battery operating state;

[0019] Calculate the change rate of the subsequence to capture the dynamic change characteristics of the battery operating state;

[0020] Combine the battery physical model, introduce electrochemical parameters and thermodynamic parameters, and generate an initial feature vector set containing multi-dimensional information.

[0021] In the above BMS-based battery safety management method, if there is noise interference in the initial feature vector set, the feature vectors are denoised by the wavelet transform method to obtain an optimized feature vector set, including:

[0022] Select appropriate wavelet basis functions and decomposition levels to perform multi-scale decomposition on the initial feature vector set;

[0023] Apply the threshold denoising algorithm to remove high-frequency noise at each decomposed scale;

[0024] Reconstruct the denoised feature vectors through inverse wavelet transform to obtain an optimized feature vector set.

[0025] In the above BMS-based battery safety management method, for the optimized feature vector set, the K-means clustering algorithm is used for pattern division, and the pre-established error pattern library is matched according to the clustering results to judge the distribution of potential error types, including:

[0026] Initialize the number of clustering centers of the K-means clustering algorithm and set the maximum number of iterations;

[0027] Calculate the distance between each feature vector in the optimized feature vector set and the clustering center, and classify according to the principle of the minimum distance;

[0028] Update the position of the clustering center, repeat the above steps until convergence, and obtain the final clustering result;

[0029] Match the clustering result with the patterns in the error pattern library to judge the distribution of potential error types.

[0030] In the above BMS-based battery safety management method, analyze the correlation between the time dimension and the state dimension through the decision tree classification method to determine the corresponding relationship between the abnormal cause and the operating stage, including:

[0031] Construct a decision tree model, and select the key dimensions in the feature vectors as splitting nodes;

[0032] Train the decision tree model according to the training data set to generate classification rules;

[0033] Predict the test data set using classification rules and analyze the correlation between the time dimension and the state dimension;

[0034] According to the prediction results, determine the corresponding relationship between the cause of the anomaly and the battery operation stage.

[0035] In the above battery safety management method based on BMS, a conditional probability calculation method is used to evaluate the error severity of each state point, and an error stage identification result is obtained, including:

[0036] Define the conditional probability formula and calculate the occurrence probability of each state point in different error stages;

[0037] According to the maximum value of the occurrence probability, determine the error stage to which the state point belongs;

[0038] Summarize the error stage identification results of all state points to generate an overall error stage distribution map.

[0039] According to another aspect of the present invention, a battery safety management system based on BMS is provided, which includes:

[0040] A data acquisition module for acquiring battery operation data collected in real time by sensors, including voltage, current and temperature parameters, to form a continuous data stream, and using a time series sampling method to obtain an original data set;

[0041] A data segmentation module for segmenting the continuous data stream using a sliding window technique, extracting subsequences according to a preset time interval in the time dimension, and obtaining a discretized state point set;

[0042] A feature extraction module for extracting statistical features from the discretized state point set, including mean, variance and change rate, and generating an initial feature vector set in combination with a physical model;

[0043] A denoising module for, if there is noise interference in the initial feature vector set, performing denoising processing on the feature vectors by means of wavelet transform to obtain an optimized feature vector set;

[0044] A pattern division module for performing pattern division on the optimized feature vector set using the K-means clustering algorithm, and matching a pre-established error pattern library according to the clustering results to judge the distribution of potential error types;

[0045] A correlation analysis module for analyzing the correlation between the time dimension and the state dimension by means of a decision tree classification method, and determining the corresponding relationship between the cause of the anomaly and the operation stage;

[0046] An error evaluation module for evaluating the error severity of each state point using a conditional probability calculation method to obtain an error stage identification result;

[0047] The result verification module is used to extract high-risk status points from the error phase identification results, verify the identification accuracy through comparison with historical data, and output the final safety management analysis report.

[0048] In the above battery safety management system based on BMS, the data segmentation module is used for:

[0049] Set the step size and window width of the sliding window, and divide the continuous data stream into multiple overlapping or non-overlapping time periods;

[0050] Extract subsequences within each time period, and map the subsequences to a multi-dimensional feature space to form a discretized set of status points;

[0051] Among them, the step size and window width of the sliding window are dynamically adjusted according to the battery operation characteristics and data sampling frequency.

[0052] In the above battery safety management system based on BMS, the feature extraction module is used for:

[0053] Calculate the mean and variance of each subsequence as the basic statistical features characterizing the battery operation status;

[0054] Calculate the change rate of the subsequence to capture the dynamic change characteristics of the battery operation status;

[0055] Combined with the battery physical model, introduce electrochemical parameters and thermodynamic parameters to generate an initial feature vector set containing multi-dimensional information.

[0056] In the above battery safety management system based on BMS, the mode division module is used for:

[0057] Initialize the number of clustering centers of the K-means clustering algorithm and set the maximum number of iterations;

[0058] Calculate the distance between each feature vector in the optimized feature vector set and the clustering center, and classify according to the principle of the minimum distance;

[0059] Update the position of the clustering center, repeat the above steps until convergence, and obtain the final clustering result;

[0060] Match the clustering result with the patterns in the error mode library to judge the distribution of potential error types.

[0061] Advantages of the present invention: The invention realizes comprehensive monitoring and abnormal early warning of the battery operation state by collecting battery operation data in real time, using the sliding window technology to segment continuous data streams, extracting discrete state point sets, extracting features from them and performing denoising processing to obtain an optimized feature vector set, using a clustering algorithm for pattern division, matching an error pattern library to judge potential error types, combining decision tree analysis to analyze the correlation between time and state dimensions, determining the corresponding relationship between abnormal causes and operation stages, calculating the error severity through conditional probability, identifying the error stage, finally extracting high-risk state points, verifying the accuracy by comparing with historical data, and outputting a safety management analysis report. The invention effectively improves the safety and reliability of battery use and provides an important basis for the optimization of the battery management system. Brief Description of the Drawings

[0062] The present invention will be further described with reference to the accompanying drawings. However, the embodiments in the drawings do not constitute any limitation to the present invention. For those of ordinary skill in the art, other drawings can be obtained based on the following drawings without creative efforts.

[0063] Figure 1 is the flowchart of the present invention. Detailed Embodiments

[0064] The present invention will be further described in conjunction with the following embodiments.

[0065] It can be seen from Figure 1 that the embodiments of the present invention provide a battery safety management method and system based on BMS, which realizes comprehensive monitoring and safety management of the battery operation state through the collaborative work of multiple modules. In combination with Figure 1 the schematic diagram of the module structure shown, the specific embodiments of each module and their connection relationships, positional relationships, and cooperation relationships will be described in detail below.

[0066] As Figure 1As shown in the figure, the system includes a data acquisition module, a data segmentation module, a feature extraction module, a denoising module, a pattern classification module, a correlation analysis module, an error evaluation module, and a result verification module. These modules are connected in sequence according to the logical order to form a complete data processing chain. The data acquisition module is responsible for obtaining battery operation data from sensors and transmitting it to the data segmentation module; the data segmentation module segments the continuous data stream and outputs a discretized state point set to the feature extraction module; the feature extraction module generates an initial feature vector set and transmits it to the denoising module; the denoising module denoises the feature vectors and outputs an optimized feature vector set to the pattern classification module; the pattern classification module completes clustering analysis and transmits the results to the correlation analysis module; the correlation analysis module further analyzes the correlation between the time dimension and the state dimension and transmits the results to the error evaluation module; the error evaluation module calculates the error severity of each state point and transmits the results to the result verification module; finally, the result verification module generates a safety management analysis report.

[0067] In the specific implementation process, the data acquisition module collects voltage, current, and temperature parameters during the battery operation in real time through voltage sensors, current sensors, and temperature sensors installed on the battery. These sensors are connected to the data acquisition module through signal lines to ensure that the collected data can be directly transmitted to the module. The data acquisition module is internally equipped with a time series sampling unit, which is used to sample the collected continuous data stream at a preset time interval to generate an original data set. To ensure the sampling accuracy, the sampling frequency of the time series sampling unit is dynamically adjusted according to the working characteristics and application scenarios of the battery. For example, in a high-power charge and discharge scenario, the sampling frequency is set to 100 times per second; while in a low-power or static scenario, the sampling frequency can be reduced to 10 times per second. This dynamic adjustment mechanism ensures the efficiency and accuracy of data acquisition.

[0068] The data segmentation module receives the original data set from the data acquisition module and uses the sliding window technique to segment the continuous data stream. The step size and window width of the sliding window are the core parameters of this module, and their settings directly affect the accuracy and efficiency of subsequent data analysis. In practical applications, the step size of the sliding window is usually set to half of the window width to ensure that the overlapping characteristics of data segmentation can capture subtle changes in the time dimension. For example, when the window width is set to 10 seconds, the step size is 5 seconds. The segmented subsequences are mapped to a multi-dimensional feature space to form a discretized state point set. This process is completed by an embedded processor, and the processor is internally equipped with a dedicated mathematical operation unit for performing subsequence extraction and mapping operations. In addition, the data segmentation module also has a dynamic adjustment function, which can modify the parameters of the sliding window in real time according to the battery operation characteristics and data sampling frequency to adapt to different operating environments.

[0069] The feature extraction module receives the discretized state point set from the data segmentation module and extracts statistical features from it. Specifically, this module first calculates the mean and variance of each subsequence as the basic statistical features characterizing the battery operating state. The mean reflects the overall level of the subsequence, while the variance reflects the degree of fluctuation of the subsequence. On this basis, the feature extraction module further calculates the change rate of the subsequence to capture the dynamic change characteristics of the battery operating state. The calculation formula for the change rate is the difference between the end value and the start value of the current subsequence divided by the time interval. To enhance the information content of the feature vector, the feature extraction module also introduces electrochemical parameters and thermodynamic parameters in combination with the battery physical model to generate an initial feature vector set containing multi-dimensional information. These parameters include the internal resistance, open circuit voltage, and thermal conductivity of the battery, etc., which are calculated and integrated through the built-in physical modeling unit.

[0070] The denoising module receives the initial feature vector set from the feature extraction module and performs denoising processing on it. Since the initial feature vector set may contain high-frequency noise interference, which will affect the accuracy of subsequent analysis, the denoising module uses the wavelet transform method to denoise the feature vector. The specific steps are as follows: First, select an appropriate wavelet basis function and decomposition level, and perform multi-scale decomposition on the initial feature vector set; then apply the threshold denoising algorithm to remove high-frequency noise at each decomposed scale; finally, reconstruct the denoised feature vector through inverse wavelet transform to obtain the optimized feature vector set. The denoising module is internally equipped with a wavelet transform processor and a threshold calculation unit, which are used to perform multi-scale decomposition and threshold denoising operations respectively. The selection of the wavelet basis function is determined according to the spectral characteristics of the battery operating data, and the decomposition level is dynamically adjusted according to the noise intensity. For example, in the case of strong noise, the decomposition level is set to 5 layers; while in the case of weak noise, the decomposition level can be reduced to 3 layers.

[0071] The pattern division module receives the optimized feature vector set from the denoising module and uses the K-means clustering algorithm to perform pattern division on it. This module first initializes the number of cluster centers and sets the maximum number of iterations. The setting of the number of cluster centers is determined according to the complexity of the battery operating state, and usually ranges from 3 to 10. Subsequently, the pattern division module calculates the distance between each feature vector in the optimized feature vector set and the cluster centers, and classifies them according to the principle of the smallest distance. After classification, update the positions of the cluster centers and repeat the above steps until convergence to obtain the final clustering result. The clustering result is matched with the pre-established error pattern library to judge the distribution of potential error types. The error pattern library is stored in the internal memory of the pattern division module and contains various typical error patterns and their corresponding feature vector distributions. The matching process is completed by the similarity calculation unit, which calculates the similarity according to the Euclidean distance between the feature vectors and takes the pattern with the highest similarity as the matching result.

[0072] The correlation analysis module receives the clustering results from the pattern division module and analyzes the correlation between the time dimension and the state dimension through the decision tree classification method. This module first constructs a decision tree model and selects the key dimensions in the feature vector as the splitting nodes. The selection of the key dimensions is determined based on the importance scores of the features, and the scores are calculated by the built-in feature selection unit. Subsequently, the correlation analysis module uses the training data set to train the decision tree model and generates classification rules. The training data set is stored in the external database of the module and contains a large amount of historical operation data and their corresponding annotation information. After training, the classification rules are used to predict the test data set to analyze the correlation between the time dimension and the state dimension. The prediction results are output through the correlation analysis unit and are used to determine the corresponding relationship between the abnormal causes and the battery operation stages.

[0073] The error evaluation module receives the prediction results from the correlation analysis module and evaluates the error severity of each state point using the conditional probability calculation method. This module first defines the conditional probability formula and calculates the occurrence probability of each state point in different error stages. The calculation of the conditional probability formula is based on Bayes' theorem and is completed by the built-in probability calculation unit. Subsequently, the error stage to which the state point belongs is determined according to the maximum occurrence probability. The error stage identification results of all state points are summarized to generate an overall error stage distribution map, and the distribution map is displayed through the visualization unit. The error evaluation module also has a dynamic adjustment function and can update the conditional probability formula in real time according to new data inputs to improve the evaluation accuracy.

[0074] The result verification module receives the overall error stage distribution map from the error evaluation module and extracts the high-risk state points from it. The extraction of the high-risk state points is determined based on the error severity scores, and the scores are calculated by the built-in risk assessment unit. After extraction, the result verification module verifies the identification accuracy through historical data comparison. The historical data is stored in the external database of the module and contains a large number of known error cases and their corresponding annotation information. The verification process is completed by the similarity calculation unit, which calculates the similarity based on the difference in the feature vectors between the state points and uses the case with the highest similarity as the verification result. Finally, the result verification module generates a safety management analysis report and transmits it to the user terminal through the output interface. The report content includes the error stage distribution map, the list of high-risk state points, and their corresponding error types and severity scores.

[0075] The above modules achieve interconnection and interoperability through a data bus to ensure efficient data transmission between modules. The data bus adopts a high-speed serial communication protocol and supports multi-channel concurrent transmission to meet the requirements of large-scale data processing. In addition, a synchronization control unit is provided between the modules to coordinate the operation rhythm between modules and avoid data loss or delay. The operation process of the entire system is uniformly scheduled by the main control unit, which dynamically allocates computing resources according to the preset task priorities to ensure the stable operation of the system in complex dynamic scenarios. To enable relevant personnel in the technical field to better understand and implement the present invention, the specific implementation principles of the present invention are further supplemented and explained below in combination with a specific application scenario.

[0076] In practical applications, taking the battery pack of a new energy vehicle as an example, its operating environment is complex and changeable, and it may face scenarios such as high temperature, high load, and frequent charging and discharging. At this time, the battery safety management system based on BMS can achieve refined monitoring and safety management of the battery operating state through the collaborative work of multiple modules.

[0077] First, the data acquisition module collects operating parameters in real time through voltage sensors, current sensors, and temperature sensors installed on the battery pack. The data collected by these sensors is transmitted to the data acquisition module via signal lines, and the time series sampling unit inside the module dynamically adjusts the sampling frequency according to the current operating scenario. For example, when the vehicle is driving at high speed, the battery is in a high-power discharge state, and the sampling frequency is set to 100 times per second; while when the vehicle is driving at low speed or parked, the sampling frequency is reduced to 10 times per second. This dynamic adjustment mechanism ensures the efficiency of data acquisition and avoids the generation of redundant data.

[0078] Subsequently, the data segmentation module receives the original data set from the data acquisition module and uses the sliding window technique to segment the continuous data stream. In this scenario, the step size of the sliding window is set to 5 seconds, and the window width is 10 seconds. This parameter setting enables the segmented subsequences to maintain a certain overlapping characteristic in the time dimension, thereby capturing the subtle changes in the battery operating state. The segmented subsequences are mapped to a multi-dimensional feature space to form a discretized state point set. The mathematical operation unit in the embedded processor is responsible for completing this process to ensure the accuracy and real-time performance of the segmentation operation.

[0079] Next, the feature extraction module extracts statistical features from the discretized state point set. Specifically, for each subsequence, the module calculates its mean and variance, which respectively reflect the overall level and fluctuation degree of the subsequence. In addition, the module also calculates the change rate of the subsequence to capture the dynamic change characteristics of the battery operating state. On this basis, electrochemical parameters and thermodynamic parameters are introduced in combination with the battery physical model to generate an initial feature vector set containing multi-dimensional information. For example, for the operating data in a high-temperature environment, the module will focus on analyzing the change trends of the battery internal resistance and thermal conductivity to more accurately characterize the battery state.

[0080] After receiving the initial feature vector set, the denoising module performs denoising processing on it using the wavelet transform method. Since the sensor data in a high-temperature environment may be interfered by high-frequency noise, the module selects appropriate wavelet basis functions and decomposition levels to perform multi-scale decomposition on the feature vectors. In this scenario, the decomposition level is set to 5 layers to effectively remove high-frequency noise. Subsequently, the optimized feature vector set is reconstructed through the inverse wavelet transform to ensure the accuracy of subsequent analysis.

[0081] After receiving the optimized feature vector set, the pattern partitioning module performs pattern partitioning on it using the K-means clustering algorithm. In this scenario, the number of cluster centers is initially set to 5. The module calculates the distance between each feature vector and the cluster centers and classifies them according to the principle of the minimum distance. After classification, the positions of the cluster centers are updated and the above steps are repeated until convergence to obtain the final clustering result. The clustering result is matched with the patterns in the error pattern library to judge the distribution of potential error types. For example, if a certain clustering result is highly similar to the overheating pattern, it is determined that the battery may have an overheating risk during this time period.

[0082] After receiving the clustering result, the correlation analysis module analyzes the correlation between the time dimension and the state dimension through the decision tree classification method. The module constructs a decision tree model, selects the key dimensions in the feature vectors as the splitting nodes, and trains the model using the training data set. In this scenario, the module analyzes and finds that the abnormal states in a high-temperature environment are mainly related to the increase in battery internal resistance and the decrease in thermal conductivity, so as to determine the corresponding relationship between the abnormal causes and the operating stages.

[0083] After receiving the prediction result, the error evaluation module evaluates the error severity of each state point using the conditional probability calculation method. The module defines the conditional probability formula and calculates the occurrence probability of each state point in different error stages. For example, the occurrence probability of a certain state point in the overheating stage is 0.85, which is significantly higher than other stages, so it is determined that this state point belongs to the overheating stage. After summarizing the error stage identification results of all state points, an overall error stage distribution map is generated and displayed through the visualization unit.

[0084] Finally, the result verification module receives the overall error phase distribution diagram and extracts high-risk status points from it. The module verifies the recognition accuracy through comparison with historical data and generates a safety management analysis report. The report content includes the error phase distribution diagram, the list of high-risk status points, their corresponding error types, and severity scores. For example, the report points out that in a high-temperature environment, the time period when the battery pack has an overheating risk mainly concentrates on the vehicle's high-speed driving phase, and it is recommended to take cooling measures to reduce the risk.

[0085] The above modules achieve interconnection and interoperability through the data bus. The main control unit dynamically allocates computing resources according to the preset task priorities to ensure that the system can operate stably in complex dynamic scenarios. Through the above steps, this system can effectively improve the error detection accuracy and safety management ability of the BMS in complex dynamic scenarios, providing reliable safety protection for the battery pack of new energy vehicles.

[0086] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than limiting the protection scope of the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the essence and scope of the technical solutions of the present invention.

Claims

1. A battery safety management method based on BMS, characterized in that: The following steps are involved: Acquire battery operation data collected by sensors in real time, including voltage, current and temperature parameters, to form a continuous data stream, and use time series sampling method to obtain the original data set; The sliding window technology is used to segment the continuous data stream, and the subsequences are extracted according to the preset time interval in the time dimension to obtain the discretized state point set; Extract statistical features from the discretized state point set, including mean, variance and rate of change, and generate an initial feature vector set in combination with the physical model; If there is noise interference in the initial feature vector set, the feature vector is denoised by wavelet transform method to obtain the optimized feature vector set; For the optimized feature vector set, the K-means clustering algorithm is used to divide the patterns, and the pre-established error pattern library is matched according to the clustering results to determine the distribution of potential error types; The correlation between the time dimension and the state dimension is analyzed through the decision tree classification method to determine the corresponding relationship between the abnormal cause and the operation stage; The conditional probability calculation method is used to evaluate the error severity of each state point and obtain the error stage identification result; High-risk status points are extracted from the error stage identification results, and the identification accuracy is verified by historical data comparison, and the final safety management analysis report is output.

2. A battery safety management method based on BMS according to claim 1, characterized in that: The sliding window technology is used to segment the continuous data stream, and the subsequences are extracted according to the preset time intervals in the time dimension to obtain a discretized state point set, including: Set the step size and window width of the sliding window to divide the continuous data stream into multiple overlapping or non-overlapping time periods; Extract subsequences in each time period and map the subsequences into a multidimensional feature space to form a discretized state point set; The step size and window width of the sliding window are dynamically adjusted according to the battery operation characteristics and data sampling frequency.

3. The battery safety management method based on BMS according to claim 1, characterized in that: Statistical features, including mean, variance and rate of change, are extracted from the discretized state point set, and combined with the physical model to generate an initial feature vector set, including: Calculate the mean and variance of each subsequence as the basic statistical features to characterize the battery operating status; calculate the rate of change of the subsequence to capture the dynamic change characteristics of the battery operating status; Combined with the battery physical model, electrochemical parameters and thermodynamic parameters are introduced to generate an initial feature vector set containing multi-dimensional information.

4. The battery safety management method based on BMS according to claim 1, characterized in that: If there is noise interference in the initial feature vector set, the feature vector is denoised by wavelet transform method to obtain the optimized feature vector set, including: Select appropriate wavelet basis functions and decomposition levels to perform multi-scale decomposition on the initial feature vector set; apply threshold denoising algorithm to remove high-frequency noise at each scale after decomposition; The denoised feature vectors are reconstructed through inverse wavelet transform to obtain the optimized feature vector set.

5. The battery safety management method based on BMS according to claim 1, characterized in that: For the optimized feature vector set, the K-means clustering algorithm is used for pattern division, and the pre-established error pattern library is matched according to the clustering results to determine the distribution of potential error types, including: Initialize the number of cluster centers of the K-means clustering algorithm and set the maximum number of iterations; Calculate the distance between each feature vector in the optimized feature vector set and the cluster center, and classify according to the principle of minimum distance; Update the location of the cluster center and repeat the above steps until convergence to obtain the final clustering result; The clustering results are matched with the patterns in the error pattern library to determine the distribution of potential error types.

6. The battery safety management method based on BMS according to claim 1, characterized in that: The correlation between the time dimension and the state dimension is analyzed by the decision tree classification method to determine the correspondence between the abnormal cause and the operation stage, including: Build a decision tree model and select the key dimensions in the feature vector as split nodes; Train the decision tree model according to the training data set to generate classification rules; Use classification rules to predict the test data set and analyze the correlation between the time dimension and the state dimension; Based on the prediction results, the correspondence between the abnormal cause and the battery operation stage is determined.

7. A battery safety management system based on BMS, characterized by: Includes the following modules: The data acquisition module is used to obtain the battery operation data collected by the sensor in real time, including voltage, current and temperature parameters, to form a continuous data stream, and to obtain the original data set using a time series sampling method; The data segmentation module is used to segment the continuous data stream using the sliding window technology, extract subsequences according to the preset time interval in the time dimension, and obtain a discretized state point set; The feature extraction module is used to extract statistical features from the discretized state point set, including mean, variance and rate of change, and generate an initial feature vector set in combination with the physical model; A denoising module is used to denoise the feature vectors by wavelet transform method if there is noise interference in the initial feature vector set, so as to obtain an optimized feature vector set; The pattern division module is used to perform pattern division based on the optimized feature vector set using the K-means clustering algorithm, and match the pre-established error pattern library based on the clustering results to determine the distribution of potential error types; The correlation analysis module is used to analyze the correlation between the time dimension and the state dimension through the decision tree classification method to determine the corresponding relationship between the abnormal cause and the operation stage; The error assessment module is used to assess the error severity of each state point using a conditional probability calculation method to obtain an error stage identification result; The result verification module is used to extract high-risk state points from the error stage recognition results, verify the recognition accuracy through historical data comparison, and output the final safety management analysis report.

8. A battery safety management system based on BMS according to claim 7, characterized in that: The data segmentation module is used for: Set the step size and window width of the sliding window to divide the continuous data stream into multiple overlapping or non-overlapping time periods; Extract subsequences in each time period and map the subsequences into a multidimensional feature space to form a discretized state point set; The step size and window width of the sliding window are dynamically adjusted according to the battery operation characteristics and data sampling frequency.

9. A battery safety management system based on BMS according to claim 7, characterized in that: The feature extraction module is used for: Calculate the mean and variance of each subsequence as the basic statistical features to characterize the battery operating status; Calculate the rate of change of the subsequence to capture the dynamic change characteristics of the battery operating state; Combined with the battery physical model, electrochemical parameters and thermodynamic parameters are introduced to generate an initial feature vector set containing multi-dimensional information.

10. A battery safety management system based on BMS according to claim 7, characterized in that: The mode division module is used for: Initialize the number of cluster centers of the K-means clustering algorithm and set the maximum number of iterations; Calculate the distance between each feature vector in the optimized feature vector set and the cluster center, and classify according to the principle of minimum distance; Update the location of the cluster center and repeat the above steps until convergence to obtain the final clustering result; The clustering results are matched with the patterns in the error pattern library to determine the distribution of potential error types.

Images