A power battery charging fault identification and early warning method, system, device and medium

Abstract

The application discloses a power battery charging fault identification and early warning method, system, device and medium, the method comprises the following steps: collecting battery parameters, and preprocessing the battery parameter data; extracting the features of the preprocessed battery parameters; clustering the features of the preprocessed battery parameters to obtain a clustering result; obtaining standard charging features according to the clustering result; performing feature matching on the standard charging features and real-time charging features, performing fault identification and fault alarm according to the matching result; calculating the charging safety margin in real time, and judging whether the charging safety margin exceeds a threshold value, and issuing a warning according to the judgment result. Through the synergistic effect of the calibration features, real-time diagnosis and long-term early warning, the application can more comprehensively and accurately realize the early warning of battery faults, and improves the accuracy of the early warning.

Description

Technical Field

[0001] This application relates to the technical field of charging fault identification and early warning, and in particular to a method, system, device and medium for power battery charging fault identification and early warning. Background Technology

[0002] With the increasing popularity of electric vehicles, the power battery, as a core component, directly impacts the user experience. To ensure battery safety, real-time monitoring of battery health and fault warnings are necessary. While some research has been conducted on battery fault warning technology, current technologies lack comprehensive battery health assessment models, suffer from prolonged warning feedback times, and largely rely on single or limited data sources for battery status assessment. Therefore, it is essential to research a more comprehensive, accurate, and rapid battery health status warning method. Summary of the Invention

[0003] The main purpose of this application is to overcome the shortcomings and deficiencies of the prior art and provide a method, system, device and medium for identifying and warning of charging faults in power batteries. Through the synergistic effect of calibration features, real-time diagnosis and long-term warning, the system can more comprehensively and accurately predict battery faults and improve the accuracy of the warning.

[0004] To achieve the above objectives, this application adopts the following technical solution:

[0005] In a first aspect, this application provides a method for identifying and warning of charging faults in power batteries, comprising the following steps:

[0006] Collect battery parameters and preprocess the battery parameters;

[0007] Extract features from the preprocessed battery parameters;

[0008] Clustering is performed on the features of the preprocessed battery parameters to obtain clustering results;

[0009] Based on the clustering results, standard charging characteristics are obtained;

[0010] The standard charging characteristics are matched with the real-time charging characteristics, and faults are identified and alarms are triggered based on the matching results.

[0011] The system calculates the charging safety margin in real time and determines whether the charging safety margin exceeds the threshold, and issues an early warning based on the determination result.

[0012] As a preferred technical solution, the characteristics of the battery parameters include: the slope of the battery charging curve, the rate of change of battery temperature, the impedance of the battery during charging, and the energy loss during the charging process.

[0013] As a preferred technical solution, the clustering is performed using the DBSCAN clustering algorithm, and the clustering process is as follows:

[0014] Initialization: Select parameters ε and MinPts to determine the conditions for forming a new cluster and labeling points as noise; where ε is the neighborhood radius and MinPts is the minimum number of points.

[0015] Neighborhood search: For each unlabeled point P, calculate the number of points in its ε. If the number of points is greater than or equal to MinPts, start creating a new cluster; otherwise, point P is temporarily labeled as noise.

[0016] Expanding clusters: For a new cluster, explore the number of points within the neighborhood radius ε of point P again. If the neighborhood radius ε of the number of points contains at least MinPts points, add the points in the number of points to the cluster. This process continues until all points in the neighborhood radius ε of all points in the cluster have been explored.

[0017] Clustering complete: The algorithm terminates when all points have been assigned to a cluster or labeled as noise;

[0018] Each cluster represents a battery health state; points marked as noise do not belong to any cluster.

[0019] As a preferred technical solution, the standard charging characteristics are obtained based on the clustering results, specifically:

[0020] Based on the clustering results of historical battery parameters, the center point of each battery health state category in the historical battery parameters is selected as the standard charging feature of that state.

[0021] As a preferred technical solution, the step of matching the standard charging characteristics with the real-time charging characteristics and identifying faults based on the matching results is as follows:

[0022] The cosine similarity algorithm is used to match real-time charging features with standard charging features, and the similarity score between real-time charging features and standard charging features is calculated.

[0023] If the similarity score between the real-time charging characteristics and the standard charging characteristics of an unhealthy state exceeds a preset threshold, fault identification will be performed.

[0024] After fault identification, if the diagnostic results indicate that there is a health problem with the battery, the charging process will be automatically interrupted, and a fault alarm will be issued through the vehicle system and user interface.

[0025] As a preferred technical solution, the steps of real-time calculation of charging safety margin, determination of whether the charging safety margin exceeds a threshold, and issuance of an early warning based on the determination result are as follows:

[0026] The Charging Safety Margin (CSM) is calculated using the following formula:

[0027] CSM=α×V+β×I+γ×T

[0028] Where V is the battery voltage margin, I is the charging current margin, T is the battery temperature margin, α is the weight of the battery voltage margin, β is the weight of the charging current margin, and γ is the weight of the battery temperature margin.

[0029] The lightGBM model is trained using battery parameters to determine the weights of battery voltage margin, charging current margin, and battery temperature margin.

[0030] When the Charging Safety Margin (CSM) exceeds the warning threshold, different levels of warnings are issued based on the degree to which the CSM exceeds the threshold.

[0031] As a preferred technical solution, the steps for training the lightGBM model are as follows:

[0032] The collected battery parameters are divided into training and validation sets;

[0033] Configure the parameters in the lightGBM model and train the lightGBM model using the training dataset. During training, the lightGBM model learns the relationship between features and the target and adjusts its internal parameters to minimize the prediction error.

[0034] The lightGBM model evaluates the importance of features based on the relationship between the learned features and the target; based on the evaluation results of feature importance, it determines the weights of battery voltage margin, charging current margin, and battery temperature margin.

[0035] Secondly, this application provides a power battery charging fault identification and early warning system, which is applied to the power battery charging fault identification and early warning method, including a preprocessing module, a feature extraction module, a clustering module, a charging feature module, a feature matching module, and an early warning module.

[0036] The preprocessing module is used to collect battery parameters and preprocess the battery parameters.

[0037] The feature extraction module is used to extract features of the preprocessed battery parameters;

[0038] The clustering module is used to cluster the features of the preprocessed battery parameters to obtain clustering results;

[0039] The charging feature module is used to obtain standard charging features based on the clustering results;

[0040] The feature matching module is used to match the standard charging features with the real-time charging features, identify faults and issue fault alarms based on the matching results.

[0041] The early warning module is used to calculate the charging safety margin in real time, determine whether the charging safety margin exceeds the threshold, and issue an early warning based on the determination result.

[0042] Thirdly, this application provides an electronic device, the electronic device comprising:

[0043] At least one processor; and,

[0044] A memory communicatively connected to the at least one processor; wherein,

[0045] The memory stores computer program instructions that can be executed by the at least one processor, which are then executed by the at least one processor to enable the at least one processor to perform the aforementioned method for identifying and warning of charging faults in a power battery.

[0046] Fourthly, this application provides a computer-readable storage medium storing a program, which, when executed by a processor, implements the aforementioned method for identifying and warning of charging faults in a power battery.

[0047] In summary, compared with the prior art, the effective effects of the technical solution provided in this application include at least the following:

[0048] This application proposes a method for identifying and warning of charging faults in power batteries. The method includes collecting battery parameters and preprocessing these parameters, which include historical and real-time parameters. Features of the preprocessed battery parameters are extracted. The preprocessed parameters and features are clustered to obtain clustering results. Standard charging features and real-time charging features are obtained based on the clustering results. The standard charging features and real-time charging features are matched, and fault identification and alarms are triggered based on the matching results. Through the synergistic effect of calibrated feature curves, real-time diagnosis, and long-term warnings, battery fault warnings can be achieved more comprehensively and accurately, improving both the accuracy and robustness of the system. Secondly, by matching the charging feature data of real-time battery parameters with pre-calibrated health status features, warnings can be triggered immediately when abnormal battery conditions occur, significantly shortening the fault diagnosis time. Furthermore, this application uses charging safety margins for warnings, which not only comprehensively assesses the battery's health status but also predicts the types and severity of potential future faults, allowing for appropriate charging adjustments or other preventative measures in advance, greatly improving battery safety. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 A flowchart illustrating a method for identifying and warning of charging faults in a power battery, as provided in one embodiment of this application;

[0051] Figure 2 A block diagram illustrating a method for identifying and warning of charging faults in a power battery, provided as an embodiment of this application;

[0052] Figure 3 This is a block diagram of a power battery charging fault identification and early warning system provided in one embodiment of this application. Detailed Implementation

[0053] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative effort are within the scope of protection of the present application.

[0054] In this application, the reference to "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described in this application can be combined with other embodiments.

[0055] Please see Figure 1 One embodiment of this application provides a method for identifying and warning of charging faults in a power battery, comprising the following steps:

[0056] Please see Figure 2 The power battery charging fault identification and early warning method in this embodiment is divided into a charging characteristic calibration part, a real-time monitoring and diagnosis part, and an early warning part;

[0057] The charging characteristic calibration section is used to establish the battery state baseline; the real-time monitoring and diagnosis section is responsible for real-time battery state monitoring and diagnosis; the early warning section is used to predict and calculate the current battery charging safety margin based on current and historical data, and issue early warnings.

[0058] S1. Collect battery parameters and preprocess the battery parameters;

[0059] The battery parameters include historical battery parameters and real-time battery parameters;

[0060] Furthermore, the collection of battery parameters involves initiating a battery charging test, charging battery packs of different years, brands, and charge / discharge cycles, and collecting battery parameters such as voltage, current, and temperature in real time during the charging process.

[0061] Furthermore, the preprocessing includes normalizing, denoising, and feature selection of the collected battery parameter data to ensure data quality and consistency.

[0062] S2. Extract the features of the preprocessed battery parameters;

[0063] Specifically, in the feature extraction process, the following key features are mainly extracted:

[0064] (1) The slope of the battery charging curve: This can reflect the battery's charging capacity and health status.

[0065] (2) Rate of change of battery temperature: A rapid rise or fall in temperature may indicate a potential problem;

[0066] (3) Battery charging impedance: An increase in the internal impedance of the battery may indicate battery aging or other problems;

[0067] (4) Energy loss during charging: can be used to estimate the overall efficiency of the battery;

[0068] Features are extracted through waveform and spectrum analysis to more accurately capture various aspects of battery status.

[0069] S3. Cluster the feature data of the preprocessed battery parameters to obtain the clustering results;

[0070] Furthermore, the features of the preprocessed battery parameters are clustered and classified using DBSCAN. DBSCAN (Density-Based Spatial Clustering of Applications with Noise) is a density-based clustering method, and the specific clustering process is as follows:

[0071] S31. Initialization: Select two parameters, neighborhood radius ε and minimum number of points MinPts, to determine when to form a new cluster and when to label points as noise.

[0072] S32. Neighborhood Search: For each unlabeled point P, calculate the number of points in its neighborhood radius ε. If this number is greater than or equal to MinPts, start creating a new cluster; otherwise, point P is temporarily labeled as noise.

[0073] S33. Expanding Clustering: For a new cluster, explore all points within the neighborhood radius ε of point P again. If the neighborhood radius ε of these points also contains at least MinPts points, then these points are also added to the cluster. This process continues until all points within the neighborhood radius ε of all points in the cluster have been explored.

[0074] S34. Clustering complete: The algorithm ends when all points have been assigned to a specific cluster or labeled as noise.

[0075] Specifically, in this embodiment, each resulting cluster represents a typical battery health state; points marked as noise do not belong to any cluster, and they may represent some abnormal battery behavior or occasional events during testing.

[0076] S4. Based on the clustering results, standard charging characteristics are obtained;

[0077] Specifically, based on the clustering results of historical battery parameters, the center point of each battery health status category is selected as the standard charging feature of that status, and the standard charging feature is labeled with the corresponding category label.

[0078] S5. Perform feature matching between the standard charging characteristics and the real-time charging characteristics, identify faults based on the matching results, and issue a fault alarm; the steps are as follows:

[0079] S51. Use the cosine similarity algorithm to match real-time data features with pre-calibrated standard charging features. Calculate the similarity score between the real-time data and each standard feature curve;

[0080] S52. If the similarity score between real-time data and the standard characteristic curve of an unhealthy state exceeds a preset threshold, the fault identification submodule will trigger fault identification.

[0081] S53 Alarm Submodule: After fault identification, if the diagnostic results indicate that the battery has obvious health problems, the module will automatically interrupt the charging process and issue a fault alarm through the vehicle system and user interface.

[0082] In this embodiment, the execution process of generating real-time charging features is the same as that of generating standard charging features. That is, the real-time battery parameters are collected first, then the real-time battery parameters are preprocessed, then the real-time battery parameter features are extracted, and finally clustering is performed based on the real-time battery parameters and features to obtain the clustering results of the real-time battery parameters. Based on the clustering results, the real-time charging features are obtained.

[0083] S6. Perform feature matching between the standard charging characteristics and the real-time charging characteristics, identify faults based on the matching results, and issue a fault alarm; the steps are as follows:

[0084] S61. Calculate the Charging Safety Margin (CSM) using the following formula:

[0085] CSM = α×V + β×I + γ×T;

[0086] Where V is the battery voltage margin, I is the charging current margin, T is the battery temperature margin, α is the weight of the battery voltage margin, β is the weight of the charging current margin, and γ is the weight of the battery temperature margin.

[0087] S62. Train the lightGBM model using battery parameters to determine the weights of battery voltage margin, charging current margin, and battery temperature margin.

[0088] Specifically, the steps for training and weight optimization of the lightGBM model are as follows:

[0089] S62.1 Data Splitting: The dataset will be split into training and validation sets to ensure that the lightGBM model performs well on unseen data.

[0090] S62.2 Model Parameter Settings: Select appropriate parameters to configure the model, such as learning rate, tree depth, minimum data size of leaf nodes, etc. These parameters will affect the performance and training speed of the lightGBM model.

[0091] S62.3 LightGBM Model Training: Using the training dataset, begin training the lightGBM model. During training, the lightGBM model learns the relationship between features and the target and adjusts its internal parameters to minimize prediction error.

[0092] S62.4 Feature Importance Assessment: The lightGBM model provides a feature importance assessment function, which can evaluate which features play a key role in the decision-making of the lightGBM model. This step is crucial for obtaining the weights α, β, and γ.

[0093] S62.5 Weight Determination: Based on the results of the feature importance assessment, the weight of each feature can be determined; for example, if the feature importance of voltage is the highest, then the value of α will be the largest, and so on.

[0094] Finally, it is worth noting that the weights are not static. As time goes on and new data is collected, the model needs to be retrained periodically and the weights α, β, γ updated to ensure that the model's predictions remain accurate and useful.

[0095] S63. When the charging safety margin (CSM) exceeds the warning threshold, different levels of warnings are issued according to the degree to which the charging safety margin (CSM) exceeds the threshold, such as: minor, moderate and severe. After the user and system notification module issue the warning, the user is notified of the charging risk level of the battery through the APP or other means.

[0096] In this embodiment, the Charging Safety Margin (CSM) is a comprehensive indicator that combines multiple factors related to battery health. The weights α, β, and γ are optimized using machine learning methods, and these weights reflect the degree of influence of each factor on battery health.

[0097] In summary, this application proposes a novel architecture that integrates feature extraction, accurate identification, and charging safety margin calculation. Unlike most existing technologies that focus on a single model or a single data source, this solution achieves more comprehensive and accurate battery fault warnings through the synergistic effect of calibrated features, real-time diagnosis, and long-term early warning. This not only improves the accuracy of the warnings but also increases the robustness of the system.

[0098] Secondly, this application achieves rapid alarm through feature matching. Specifically, by matching real-time battery data with pre-calibrated health status feature curves, an early warning can be triggered immediately when the battery status is abnormal, significantly shortening the fault diagnosis time.

[0099] Finally, this application introduces the concept of "charging safety margin" for the first time. This innovative concept can not only more comprehensively assess the health status of the battery, but also predict the types and severity of future failures, thereby enabling appropriate charging adjustments or other preventive measures to be taken in advance, which greatly improves the safety of battery use.

[0100] It should be noted that, for the sake of simplicity, the aforementioned method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously.

[0101] Based on the same idea as the power battery charging fault identification and early warning method in the above embodiments, this application also provides a power battery charging fault identification and early warning system, which can be used to execute the above-described power battery charging fault identification and early warning method. For ease of explanation, the schematic diagram of an embodiment of a power battery charging fault identification and early warning device only shows the parts related to the embodiments of this application. Those skilled in the art will understand that the illustrated structure does not constitute a limitation on the device, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0102] Please see Figure 3 In another embodiment of this application, a power battery charging fault identification and early warning system 100 is provided. The system includes a preprocessing module 101, a feature extraction module 102, a clustering module 103, a charging feature module 104, a feature matching module 105, and an early warning module 106.

[0103] The preprocessing module 101 is used to collect battery parameters and preprocess the battery parameters.

[0104] The feature extraction module 102 is used to extract features of the preprocessed battery parameters;

[0105] The clustering module 103 is used to cluster the features of the preprocessed battery parameters to obtain clustering results;

[0106] The charging feature module 104 is used to obtain standard charging features based on the clustering results;

[0107] The feature matching module 105 is used to perform feature matching between the standard charging features and the real-time charging features, and to identify and alarm faults based on the matching results.

[0108] The early warning module 106 is used to calculate the charging safety margin in real time, determine whether the charging safety margin exceeds the threshold, and issue an early warning based on the determination result.

[0109] It should be noted that the power battery charging fault identification and early warning device of this application corresponds one-to-one with the power battery charging fault identification and early warning method of this application. The technical features and beneficial effects described in the above-mentioned embodiments of the power battery charging fault identification and early warning method are all applicable to the embodiments of the power battery charging fault identification and early warning system. For details, please refer to the description in the method embodiments of this application, which will not be repeated here.

[0110] Furthermore, in the above embodiment of a power battery charging fault identification and early warning system, the logical division of each program module is only an example. In actual applications, the above functions can be assigned to different program modules as needed, for example, for the sake of corresponding hardware configuration requirements or software implementation convenience. That is, the internal structure of the power battery charging fault identification and early warning system is divided into different program modules to complete all or part of the functions described above.

[0111] In another embodiment, an electronic device is provided for implementing a method for identifying and warning of power battery charging faults, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor; when the processor executes the computer program, it implements a method for identifying and warning of power battery charging faults according to any embodiment of this application.

[0112] For example, in this embodiment, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete this application. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the device.

[0113] The device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The device may include, but is not limited to, a processor and memory.

[0114] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the device, connecting various parts of the device via various interfaces and lines.

[0115] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function, etc. In addition, the memory may include high-speed random access memory and non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0116] Accordingly, this application also provides a computer-readable storage medium, which includes a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute a power battery charging fault identification and early warning method as described in any of the above embodiments.

[0117] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0118] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0119] The above embodiments are preferred embodiments of this application, but the implementation of this application is not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of this application shall be considered equivalent substitutions and shall be included within the protection scope of this application.

Claims

1. A method for identifying and warning of charging faults in power batteries, characterized in that, Includes the following steps: Collect battery parameters and preprocess the battery parameters; Extract features from the preprocessed battery parameters; The DBSCAN clustering algorithm is used to cluster the features of the preprocessed battery parameters to obtain the clustering results; The DBSCAN clustering algorithm's clustering process includes: Initialization: Selecting the neighborhood radius e and the minimum number of points MinPts; Neighborhood search: For each unlabeled point P, calculate the number of points in its neighborhood radius e. If the number of points is greater than or equal to MinPts, start creating a new cluster; otherwise, temporarily label point P as noise; Expanding the cluster: For a new cluster, explore all points within the neighborhood radius e of point P. If these points also contain at least MinPts points within their neighborhood radius e, add these points to the cluster and continue expanding until all points in the cluster have had their neighborhood radius e explored; Completing the cluster: The algorithm ends when all points have been assigned to a cluster or labeled as noise. Each cluster represents a battery health state, and points labeled as noise do not belong to any cluster. Based on the clustering results, standard charging features are obtained, specifically including: selecting the center point of each battery health state category in the historical battery parameters as the standard charging feature of that state based on the clustering results of the historical battery parameters. The standard charging characteristics are matched with the real-time charging characteristics, and fault identification and alarm are triggered based on the matching results. Specifically, a cosine similarity algorithm is used to match the real-time charging characteristics with the standard charging characteristics, and the similarity score between the real-time charging characteristics and the standard charging characteristics in an unhealthy state is calculated. If the similarity score between the real-time charging characteristics and the standard charging characteristics in an unhealthy state exceeds a preset threshold, fault identification will be performed. After fault identification, if the diagnostic results indicate that the battery has a health problem, the charging process will be automatically interrupted, and a fault alarm will be issued through the vehicle system and user interface. The system calculates the charging safety margin in real time and determines whether it exceeds a threshold. Based on the determination, it issues an early warning. Specifically, this includes: calculating the Charging Safety Margin (CSM) using the formula: CSM = α × V + β × I + γ × T, where V is the battery voltage margin, I is the charging current margin, T is the battery temperature margin, α is the weight of the battery voltage margin, β is the weight of the charging current margin, and γ is the weight of the battery temperature margin; training the lightGBM model using battery parameters to determine the weights of the battery voltage margin, charging current margin, and battery temperature margin; and issuing different levels of warnings based on the degree to which the CSM exceeds the threshold.

2. The method for identifying and warning of charging faults in a power battery according to claim 1, characterized in that, The characteristics of the battery parameters include: the slope of the battery charging curve, the rate of change of battery temperature, the impedance of the battery during charging, and the energy loss during the charging process.

3. The method for identifying and warning of charging faults in a power battery according to claim 1, characterized in that, The steps for training the lightGBM model are as follows: The collected battery parameters are divided into training and validation sets; Configure the parameters in the lightGBM model and train the lightGBM model using the training dataset. During training, the lightGBM model learns the relationship between features and the target and adjusts its internal parameters to minimize the prediction error. The lightGBM model evaluates the importance of features based on the relationship between the learned features and the target. Based on the assessment results of feature importance, the weights of battery voltage margin, charging current margin, and battery temperature margin are determined.

4. A power battery charging fault identification and early warning system, characterized in that, A power battery charging fault identification and early warning method applicable to any one of claims 1-3 includes a preprocessing module, a feature extraction module, a clustering module, a charging feature module, a feature matching module, and an early warning module; The preprocessing module is used to collect battery parameters and preprocess the battery parameters. The feature extraction module is used to extract features of the preprocessed battery parameters; The clustering module uses the DBSCAN clustering algorithm to cluster the features of the preprocessed battery parameters to obtain clustering results. The DBSCAN clustering algorithm's clustering process includes: Initialization: Selecting the neighborhood radius e and the minimum number of points MinPts; Neighborhood search: For each unlabeled point P, calculate the number of points in its neighborhood radius e. If the number of points is greater than or equal to MinPts, start creating a new cluster; otherwise, temporarily label point P as noise; Expanding the cluster: For a new cluster, explore all points within the neighborhood radius e of point P. If these points also contain at least MinPts points within their neighborhood radius e, add these points to the cluster and continue expanding until all points in the cluster have had their neighborhood radius e explored; Completing the cluster: The algorithm ends when all points have been assigned to a cluster or labeled as noise. Each cluster represents a battery health state, and points labeled as noise do not belong to any cluster. The charging feature module is used to select the center point of each battery health state category as the standard charging feature of that state based on the clustering results. The feature matching module is used to match the standard charging features with the real-time charging features, identify faults and issue fault alarms based on the matching results; wherein, a cosine similarity algorithm is used to match the real-time charging features with the standard charging features and calculate the similarity score; if the similarity score between the real-time charging features and the standard charging features in an unhealthy state exceeds a preset threshold, fault identification is performed, the charging process is automatically interrupted, and a fault alarm is issued through the vehicle system and the user interface. The early warning module is used to calculate the charging safety margin (CSM) in real time. The calculation formula is: CSM = α × V + β × I + γ × T, where V is the battery voltage margin, I is the charging current margin, T is the battery temperature margin, α is the weight of the battery voltage margin, β is the weight of the charging current margin, and γ is the weight of the battery temperature margin. The lightGBM model is trained using battery parameters, and the weights of α, β, and γ are determined by the feature importance evaluation results of the lightGBM model. When the CSM exceeds the warning threshold, different levels of warnings are issued according to the degree to which the CSM exceeds the threshold.

5. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores computer program instructions executable by the at least one processor, the computer program instructions being executed by the at least one processor to enable the at least one processor to perform a power battery charging fault identification and early warning method as described in any one of claims 1-3.

6. A computer-readable storage medium storing a program, characterized in that, When the program is executed by the processor, it implements the power battery charging fault identification and early warning method according to any one of claims 1-3.

Images