A safety warning method, device and electronic equipment for power battery

Abstract

The embodiment of the present application discloses a safety warning method, device, and electronic device for a power battery, which belongs to the field of battery management technology and can accurately diagnose power battery faults and identify faulty vehicles to achieve early warning. It includes: obtaining the operating condition data of the power battery during charging, and dividing the operating condition data into segments to obtain the voltage data corresponding to the target segment; calculating the voltage parameter index of the target segment according to the voltage data; the voltage parameter index includes: the voltage change rate difference ratio index and the voltage consistency difference index; normalizing the voltage parameter index to obtain a feature vector, which includes: the voltage change rate feature vector and the voltage consistency vector; inputting the feature vector into a pre-trained vector machine model, and outputting judgment data, which includes whether the target segment is a fault segment or a normal segment; and based on the judgment data, performing safety warning on the power battery.

Description

Technical Field

[0001] The present application relates to the field of battery management technology, and in particular to a safety warning method, device, electronic device and storage medium for a power battery. Background Art

[0002] As the core component of new energy vehicles, power batteries have numerous failure causes and complex coupling mechanisms. Therefore, accurate battery fault diagnosis and safety warning can better ensure the safe use of new energy vehicles.

[0003] Among the related technologies, the data-driven power battery fault diagnosis method has attracted the attention of many researchers. It does not require a systematic and precise analytical mathematical model, but relies on sample data to establish the connection between data features and fault characterization. It mainly analyzes operating data such as voltage and temperature, and uses signal processing and feature extraction technology to convert them into safety feature vectors. It uses machine learning algorithms for classification and diagnosis to complete power battery fault diagnosis and identify faulty vehicles for early warning. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a power battery safety warning method, device, electronic device and storage medium to solve the problem of how to accurately diagnose power battery failures and identify faulty vehicles to achieve early warning.

[0005] To solve the above technical problems, the embodiments of the present application are implemented as follows:

[0006] In a first aspect, an embodiment of the present application provides a safety warning method for a power battery, comprising: obtaining operating condition data of the power battery during charging, and dividing the operating condition data into segments to obtain voltage data corresponding to a target segment; calculating a voltage parameter index of the target segment based on the voltage data; the voltage parameter index includes a voltage change rate difference ratio index and a voltage consistency difference index; normalizing the voltage parameter index to obtain a feature vector, the feature vector includes a voltage change rate feature vector and a voltage consistency vector; inputting the feature vector into a pre-trained vector machine model, and outputting judgment data, the judgment data including whether the target segment is a fault segment or a normal segment; and performing a safety warning on the power battery based on the judgment data.

[0007] In a second aspect, an embodiment of the present application provides a safety warning device for a power battery, comprising: an acquisition module, used to acquire operating condition data of the power battery during charging, and divide the operating condition data into segments to obtain voltage data corresponding to the target segment; a calculation module, used to calculate the voltage parameter index of the target segment based on the voltage data; the voltage parameter index includes: a voltage change rate difference ratio index and a voltage consistency difference index; a processing module, used to normalize the voltage parameter index to obtain a feature vector, the feature vector includes: a voltage change rate feature vector and a voltage consistency vector; an output module, used to input the feature vector into a pre-trained vector machine model, and output judgment data, the judgment data including whether the target segment is a fault segment or a normal segment; based on the judgment data, a safety warning is performed on the power battery.

[0008] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor and a memory electrically connected to the processor, the memory storing a computer program, and the processor being used to call and execute the computer program from the memory to implement the above-mentioned power battery safety warning method.

[0009] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium for storing a computer program, wherein the computer program can be executed by a processor to implement the above-mentioned power battery safety warning method.

[0010] In a fifth aspect, an embodiment of the present application provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the above-mentioned power battery safety warning method.

[0011] By adopting the technical solution of the embodiment of the present application, the operating condition data of the power battery during charging is obtained, and the operating condition data is divided into segments to obtain voltage data corresponding to the target segment; based on the voltage data, the voltage parameter index of the target segment is calculated; the voltage parameter index includes: a voltage change rate difference ratio index and a voltage consistency difference index; the voltage parameter index is normalized to obtain a feature vector, and the feature vector includes: a voltage change rate feature vector and a voltage consistency vector; the feature vector is input into a pre-trained vector machine model, and judgment data is output, and the judgment data includes whether the target segment is a fault segment or a normal segment; based on the judgment data, a safety warning is given to the power battery. It can be seen that by obtaining the operating condition data of the power battery during the charging period and calculating the voltage parameter index corresponding to the target segment, the effective data safety index of the power battery can be obtained, and through the learning method of the support vector machine model, the voltage parameter index is input into the support vector machine model to output the judgment data. According to the judgment data, it is judged whether the target segment is a fault segment. The operating condition data of the power battery can be fully analyzed, and then the vehicle with the power battery with the faulty segment can be given a safety warning. There is no need to discover the safety problem of the vehicle when the vehicle is in use. The power battery fault can be accurately diagnosed and the faulty vehicle can be identified to achieve early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate one or more embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in one or more embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0013] Figure 1 is a schematic flow chart of a safety warning method for a power battery according to an embodiment of the present application;

[0014] Figure 2 is a schematic flow chart of a safety warning method for a power battery according to another embodiment of the present application;

[0015] Figure 3 is a schematic block diagram of a safety warning device for a power battery according to an embodiment of the present application;

[0016] Figure 4 It is a schematic diagram of the hardware structure of a safety warning device for a power battery according to an embodiment of the present application. DETAILED DESCRIPTION

[0017] The embodiments of the present application provide a power battery safety warning method, device, electronic device and storage medium to solve the problem of accurately diagnosing power battery failures and identifying faulty vehicles to achieve early warning.

[0018] In order to enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.

[0019] The power battery safety warning method provided in the embodiment of the present application can be executed by an electronic device, or by software installed in the electronic device. Specifically, the electronic device can be a terminal device or a server device. Among them, the terminal device can include a smart phone, a laptop computer, a smart wearable device, a vehicle terminal, etc., and the server device can include an independent physical server, a server cluster composed of multiple servers, or a cloud server capable of cloud computing.

[0020] In the following, in conjunction with the accompanying drawings, a safety warning method for a power battery provided in an embodiment of the present application is described in detail through specific embodiments and application scenarios.

[0021] Figure 1 A schematic flow chart of a power battery safety warning method provided by an embodiment of the present invention is shown, and the method comprises the following steps:

[0022] S102, obtaining operating condition data of the power battery during charging, and dividing the operating condition data into segments to obtain voltage data corresponding to target segments.

[0023] The actual vehicle operation data can be divided into charging condition, discharging condition and static condition according to the operating conditions. Obtain the operating condition data of the power battery during charging, including screening the operating condition data of the power battery during charging condition. The screening conditions are: ① The charging flag bit specified by the national standard is 1 or 4 ② The vehicle is stationary during charging and the speed is 0 ③ The total current of the battery pack is negative. In other words, the data that meets the above three conditions at the same time is the operating condition data of the power battery during charging.

[0024] The operating condition data is segmented, where the charging current can be divided into charging segments of different rates. The time interval between two adjacent frames of data exceeds 300s, which is considered to be a different charging segment. On this basis, the operating condition data of the power battery during charging is segmented to obtain the divided power battery charging data segments, among which the segments that need to be detected are the target segments, and the voltage data of the single cell corresponding to each frame in the target segment is obtained to form a voltage list. The single cell voltage list matrix of each frame is the voltage data corresponding to the target segment. For example, the target segment is the i-th segment, and the corresponding voltage data includes the voltage matrix V i As shown in the following formula:

[0025]

[0026] in, represents the voltage value of the jth frame of the kth single cell number in the segment, n represents the total number of single cells, and m represents the total number of frames in the segment.

[0027] It should be understood that a power battery can be understood as a battery pack, wherein the battery pack can include multiple single cells and multiple frames. The operating condition data can be divided into multiple segments, and if a target segment fails, the entire power battery is a failed power battery.

[0028] S104, calculating a voltage parameter index of the target segment according to the voltage data.

[0029] Among them, the voltage parameter indicators include: voltage change rate difference ratio indicator and voltage consistency difference indicator.

[0030] Specifically, under normal circumstances, the data consistency of the power battery pack will be very good, but as the number of charging cycles increases, the parameters of the potential faulty single cells will change. Therefore, the voltage change rate difference ratio index and the voltage consistency difference index in the voltage parameter index are used as the basis for fault diagnosis to carry out the diagnosis of power battery faults. Using the voltage change rate difference ratio index and the voltage consistency difference index as diagnostic indicators for power battery faults can improve the accuracy of diagnosing power battery faults.

[0031] According to the voltage data obtained in step 102, the voltage data of each frame of the single cell corresponding to the target segment of the power battery can be obtained, and then the voltage parameter index of the target segment can be calculated, such as the voltage change rate difference ratio index and the voltage consistency difference index corresponding to the target segment, which are used to diagnose the fault of the target segment.

[0032] S106, normalizing the voltage parameter index to obtain a characteristic vector.

[0033] The characteristic vectors include: voltage change rate characteristic vector and voltage consistency vector.

[0034] The voltage parameter index obtained in step 104 is normalized. Normalization is a data processing method, the purpose of which is to convert dimensioned data into dimensionless data expression, such as removing the dimension of the vector of the voltage parameter index to ensure that the eigenvector values ​​are between [-1, 1]. The eigenvector obtained by normalization can simplify the calculation and reduce the value, facilitate subsequent data processing, and ensure faster convergence when the program is running.

[0035] S108, inputting the feature vector into a pre-trained vector machine model, outputting judgment data, the judgment data including whether the target segment is a faulty segment or a normal segment; and providing a safety warning for the power battery based on the judgment data.

[0036] According to inputting the feature vector into the pre-trained vector machine model, the vector machine model is a pre-trained vector machine model of the support vector machine (SVM) algorithm, namely, the SVM model. The vector machine model is a generalized linear classifier widely used in statistical classification and regression analysis. By inputting the feature vector into the pre-trained vector machine model, the feature vector can be classified and processed, and whether the target segment corresponding to the feature vector is faulty is identified, and judgment data is output. According to the judgment data, it is determined whether the target segment is a faulty segment or a normal segment.

[0037] Based on the judgment data, a safety warning is issued for the power battery. Specifically, when the judgment data identifies that the target is a faulty segment, the power battery corresponding to the target segment is faulty, and the vehicle using the power battery is also judged as a faulty vehicle; when the judgment data identifies that the target is a normal segment, the power battery corresponding to the target segment is normal, and the vehicle using the power battery is also judged as a normal vehicle. It is possible to determine whether the target segment is a faulty segment based on the output judgment data. If it is a faulty segment, a safety warning is issued for the power battery, and the corresponding vehicle is determined to be a faulty vehicle in advance, without having to determine that the vehicle has a safety problem when the vehicle is in use.

[0038] By adopting the technical solution of the embodiment of the present application, the operating condition data of the power battery during charging is obtained, and the operating condition data is divided into segments to obtain voltage data corresponding to the target segment; based on the voltage data, the voltage parameter index of the target segment is calculated; the voltage parameter index includes: a voltage change rate difference ratio index and a voltage consistency difference index; the voltage parameter index is normalized to obtain a feature vector, and the feature vector includes: a voltage change rate feature vector and a voltage consistency vector; the feature vector is input into a pre-trained vector machine model, and judgment data is output, and the judgment data includes whether the target segment is a fault segment or a normal segment; based on the judgment data, a safety warning is given to the power battery. It can be seen that by obtaining the operating condition data of the power battery during the charging period and calculating the voltage parameter index corresponding to the target segment, the effective data safety index of the power battery can be obtained, and through the learning method of the support vector machine model, the voltage parameter index is input into the support vector machine model to output the judgment data. According to the judgment data, it is judged whether the target segment is a fault segment. The operating condition data of the power battery can be fully analyzed, and then the vehicle with the power battery with the faulty segment can be given a safety warning. There is no need to discover the safety problem of the vehicle when the vehicle is in use. The power battery fault can be accurately diagnosed and the faulty vehicle can be identified to achieve early warning.

[0039] In one embodiment, the voltage change rate difference ratio index of the target segment is calculated according to the voltage data (ie, S104), and the following steps A1-A3 may be performed:

[0040] Step A1: Determine the voltage difference between the target single cell in the power battery in the target frame and the next frame of the target segment according to the voltage data, and determine the average voltage value of each single cell corresponding to the target frame and the average voltage value of each single cell corresponding to the next frame.

[0041] The power battery includes a plurality of single cells. The target segment may include a plurality of frames and a plurality of single cells, and the target frame may also include a plurality of single cells, and the number is not specifically limited here.

[0042] The voltage data is voltage data of a target segment of the power battery under charging conditions in step 102 . The power battery is a battery pack including a plurality of single cells. The voltage data includes a plurality of single cells and a plurality of frames.

[0043] The voltage difference between the target single cell in the power battery in the target frame of the target segment and the next frame, using the example of step 102, the target segment is i; the target single cell in the power battery in the target frame of the target segment can be k, and the target frame is j. Then the voltage value of the target single cell in the target frame of the target segment can be expressed as The voltage value of the target single cell in the next frame of the target frame of the target segment can be expressed as The voltage difference between the target single cell in the target frame and the next frame of the target segment The voltage value of the target single cell in the next frame of the target frame of the target segment minus the voltage value of the target single cell in the target frame of the target segment is expressed as:

[0044]

[0045] The average voltage value of each single cell corresponding to the target frame. It should be understood that the target frame includes multiple single cells, and the average voltage value of all single cells in the target frame is calculated. For example, using the example of step 102, the target segment is i, the target frame is j, and the average voltage value of each single cell corresponding to the target frame is Then, the average voltage value of each single cell corresponding to the next frame can be expressed as

[0046] Step A2: Determine the average voltage difference corresponding to the target frame according to the average voltage value of each single cell corresponding to the target frame and the average voltage value of each single cell corresponding to the next frame.

[0047] Specifically, according to the average voltage value of each single battery corresponding to the target frame obtained in step A1 The average voltage value of each single cell corresponding to the next frame Calculate the average voltage difference corresponding to the target frame, which can be expressed as:

[0048]

[0049] Step A3: Based on the voltage difference and the average voltage difference, determine the voltage change rate difference ratio of the target single cell of the target segment, and use the voltage change rate difference ratio of the target single cell as the voltage change rate difference ratio indicator of the target segment.

[0050] According to step A2, the voltage difference of the target single cell in the target segment is The average voltage difference of all cells corresponding to the target frame in the target segment is Then, the voltage change rate difference ratio of the target single battery in the target segment is It can be calculated according to the following formula:

[0051]

[0052] The voltage change rate difference of the target single cell in the target segment is The voltage change rate difference ratio index of the target segment is determined.

[0053] In this embodiment, the voltage difference between the target single cell in the power battery in the target segment and the next frame is calculated based on the voltage data, and the average voltage difference corresponding to the target frame in the target segment is determined based on the calculated voltage average of each single cell in the target frame of the target segment and the corresponding voltage average of each single cell in the next frame. Based on the voltage difference between the target single cell in the target segment and the next frame, and the average voltage difference corresponding to the target frame, the voltage change rate difference ratio of each single cell in the target segment is determined, and then the voltage change rate difference ratio index of the target segment is determined. Based on the voltage change rate difference ratio index, the voltage changes of each single cell in the power battery are analyzed in detail, which can provide more accurate data for the target segment for fault analysis and diagnosis.

[0054] In one embodiment, the voltage consistency difference index of the target segment is calculated according to the voltage data (ie, S104), and the following steps B1-B2 are performed:

[0055] Step B1: Determine the voltage average value of the target frame in the target segment and the voltage value of each single cell corresponding to the target frame in the target frame according to the voltage data.

[0056] According to the voltage data, the voltage value of each single battery corresponding to the target frame in the target segment is obtained, and the voltage value of each single battery is obtained according to the voltage value of each single battery. Calculate the average voltage of each single battery corresponding to the target frame j The calculation formula is as follows:

[0057]

[0058] in, When the target segment is the i-th segment, the maximum and minimum values ​​of the voltage values ​​of each single cell in the voltage data matrix when the target frame is the j-th frame, The voltage average value of the target frame is calculated after removing the single cells with the maximum voltage value and the minimum voltage value from the single cells corresponding to the target frame. Removing the extreme value can avoid the influence of the extreme value on the overall result.

[0059] Step B2: Determine the voltage consistency difference index of the target segment based on the difference between the voltage value of each single cell corresponding to the target frame in the target frame and the voltage average value.

[0060] The voltage value of the target single cell in the target frame in the target segment is expressed as According to step B1, the voltage average value is obtained Then, the voltage consistency difference of each single cell K in each frame j under the target segment i is expressed as It is expressed as the voltage consistency difference index of the target segment, as shown in the following formula:

[0061]

[0062] In this embodiment, by calculating the voltage average value of the target segment in the target frame, further calculating the voltage value of each single cell in each target frame of the target segment, and the difference between the corresponding voltage average values ​​of each target frame, the difference represents the voltage consistency difference index of the target segment. According to the voltage consistency difference index of the target segment, it can be determined whether there is a sudden change in the voltage in the target segment, and by deeply analyzing whether the voltage in the target segment changes, the efficiency and accuracy of detecting whether the target segment is faulty can be improved.

[0063] In one embodiment, the voltage parameter index is normalized to obtain a characteristic vector (ie, S106), and the following step C may be performed:

[0064] Step C: Based on the Min-Max normalization algorithm, the voltage change rate difference ratio index and the voltage consistency difference index are normalized to obtain the corresponding feature vector.

[0065] Before using the support vector machine method, the voltage change rate difference ratio index and the voltage consistency difference index obtained in step A3 and step B2 need to be normalized to obtain the corresponding voltage change rate feature vector and voltage consistency vector. Specifically, the Min-Max normalization algorithm removes the dimension and obtains two feature vectors Z1 and Z2. It is ensured that the feature vector values ​​are between [-1, 1]. The calculation formulas of Z1 and Z2 obtained by using the Min-Max normalization algorithm are as follows:

[0066]

[0067] in, Represent characteristic parameters The minimum and maximum values ​​of the kth monomer in the ith segment.

[0068] In this embodiment, the voltage change rate difference ratio index and the voltage consistency difference index are converted into dimensionless feature vectors through normalization processing, which provides a basis for vector machine calculation and improves calculation efficiency.

[0069] In one embodiment, the training of the vector machine model may perform the following steps D1-D5:

[0070] Step D1: Obtain sample feature vectors of multiple sample power batteries when they are charged and corresponding sample fault fragments, and input the sample feature vectors into a vector machine model.

[0071] The voltage parameter index of the sample target segment corresponding to the multiple sample power batteries is obtained. The voltage parameter index includes the voltage change rate difference ratio index and the voltage consistency difference index. The sample feature vector is the voltage parameter index after normalization, including the voltage change rate feature vector and the voltage consistency vector. The sample feature vector corresponding to the multiple sample power batteries is obtained, that is, the voltage parameter index in the sample target segment corresponding to the multiple sample power batteries after normalization is obtained.

[0072] The sample fault fragments are included in the target sample fragments of obtaining multiple sample power batteries, including normal fragments of normal power batteries corresponding to normal vehicles and fault fragments of faulty power batteries corresponding to faulty vehicles. The fault fragments are used as sample fault fragments as a verification set to verify the output result after the sample feature vector is input into the vector machine model.

[0073] For example, the sample set in the input vector machine model is: D i =(x i ,y i ),i=1,2,…n,y i ∈{+1,-1}, where x i is the matrix set of sample feature vectors, y i is the sample category, and n is the total sample size.

[0074] Step D2: Obtain the hyperplane equation in the vector machine model, as well as the corresponding objective function and constraints.

[0075] The vector machine model is the SVM model of the support vector machine. The main purpose of SVM is to find an optimal hyperplane in the sample space so that the distance between the sample points on both sides of the hyperplane and the plane is maximized, thereby providing good generalization ability for classification problems. The hyperplane equation formula of the vector machine is:

[0076] ωx+b=0

[0077] In the above formula, ω is the direction control vector of the hyperplane, and b is the position control parameter of the hyperplane.

[0078] The objective function and constraints of SVM (st represents constraints) are:

[0079]

[0080] The above formula is the original SVM algorithm, which requires that all sample points must fall on both sides of the hyperplane, and the sample set to be classified is linearly separable, which is also called hard margin SVM.

[0081] In order to expand the application scope of the SVM algorithm, soft margin SVM was proposed, and its objective function and constraints are as follows:

[0082]

[0083] In the above formula, C is the penalty factor, C>0,ξ i is the slack variable, ξ i ≥0.

[0084] By using the soft margin SVM approach, a slack variable is added to the constraints to allow some points corresponding to the sample feature vectors to be wrong to a certain extent, that is, a small number of points corresponding to the sample feature vectors are allowed not to meet the constraints of the hard margin SVM, making the model more generalizable and having a wider range of applications.

[0085] Step D3: According to the kernel function in the vector machine model, the sample feature vector is mapped to a high-dimensional space.

[0086] When the vector machine model solves the nonlinear classification problem through the kernel technique, the inner product of the sample feature vector after mapping in the high-dimensional space can be expressed as a function of the corresponding sample feature vector in the original space, that is, the kernel function. The inner product of the sample feature vector after mapping in the high-dimensional space is calculated by the kernel function. The calculation formula is as follows:

[0087]

[0088] Among them, x, y are n-dimensional, f is the mapping from n-dimensional to m-dimensional,<f(x),f(y)> Represents the inner product.

[0089] Step D4: Classify the sample feature vector through the hyperplane equation and kernel function, and output the sample judgment data.

[0090] The sample feature vector can be classified through the hyperplane equation.

[0091] The kernel function maps the sample feature vector from low-dimensional space to high-dimensional space. Through mapping, the linearly inseparable sample feature vector in low-dimensional space can be transformed into linearly separable one.

[0092] The kernel function is divided into linear kernel function and Gaussian kernel function, among which the linear kernel function is:

[0093]

[0094] The linear kernel function is a special case of the polynomial kernel function. Its advantage is simplicity, but its disadvantage is that it cannot process linearly inseparable data sets. It is mainly used in linearly separable situations and has the advantages of fewer parameters and faster speed. C is the penalty coefficient. The larger the C value, the easier it is for the vector machine model to overfit, and the smaller the C value, the easier it is for the vector machine model to underfit.

[0095] The Gaussian kernel function is:

[0096]

[0097] The Gaussian kernel function, also known as the radial basis kernel function, is a kernel function with strong locality. It can expand the input sample feature vector to an infinite-dimensional space. It has good performance for both large and small samples. The default value of γ is 1 / K. The larger γ is, the fewer support vectors there are, and the faster the training and prediction speeds are.

[0098] The sample feature vector is linearly planned through the kernel function, and the sample feature vector is classified by the hyperplane equation. The vector machine model can output the classification result, that is, the sample judgment data. For example, if the sample judgment result corresponding to the sample feature vector is 1, then it is the sample normal data. If the output sample judgment result is -1, then it is the sample fault data. The sample normal data corresponds to the sample normal segment, and the sample fault data corresponds to the sample fault segment.

[0099] Step D5: Based on the sample judgment data and the sample fault fragments, adjust the kernel function and train the vector machine model.

[0100] According to the sample judgment data obtained in step D4, the sample fault fragments corresponding to the multiple sample power batteries obtained in step D1 are compared, and the selection of the kernel function is adjusted to make the output sample judgment data correspond to the sample fault fragment. The judgment data output by the sample judgment is the sample fault data, that is, the corresponding sample fault fragment. If the judgment data output by the sample judgment is the sample normal data, and the sample normal data does not correspond to the sample fault fragment, then the vector machine model training is successful.

[0101] In this embodiment, by training the vector model of the support vector machine algorithm, including classifying the sample feature data through a hyperplane and mapping the sample feature vector to a high-dimensional space for classification using a kernel function, the sample feature vector can be statistically classified to improve the accuracy and efficiency of power battery fault diagnosis.

[0102] In one embodiment, the sample judgment data and the sample fault fragment, adjusting the kernel function (ie, step D5), may perform the following steps D51-D52:

[0103] Step D51: Calculate evaluation indicators of the vector machine model based on the sample judgment data and the sample fault fragments. The evaluation indicators include one or more of the accuracy, precision, recall rate and the harmonic mean of the recall rate and precision rate of the sample judgment data.

[0104] Specifically, the number of features predicted to be positive samples and actually positive samples (True Positives, TP); the number of features predicted to be positive samples and actually negative samples (False Positives, FP); the number of features predicted to be negative samples and actually negative samples (True Positives, TN); the number of features predicted to be negative samples and actually positive samples (False Positves, FP). The evaluation indicators include one or more of the accuracy, precision, recall rate, and the harmonic mean of recall rate and precision rate of sample judgment data. The specific calculation methods are as follows:

[0105] The accuracy rate includes the percentage of correct samples to the total samples, and the calculation process is as follows:

[0106]

[0107] The precision includes the probability of actually being positive among all samples predicted to be positive for the prediction result. The calculation process is as follows:

[0108]

[0109] The recall rate includes the probability of samples predicted to be positive among the samples that are actually positive for the original samples. The calculation process is as follows:

[0110]

[0111] The calculation process of the harmonic mean of recall and precision is as follows:

[0112]

[0113] Step D52: According to the evaluation index, adjust the kernel function to a Gaussian kernel function.

[0114] According to the results calculated from the evaluation indicators, when the kernel function in the vector machine model is a Gaussian kernel function, the result output by the model is the best and most accurate. Therefore, the Gaussian kernel function can be selected as the kernel function in the vector machine model.

[0115] As an example, the linear kernel function is mainly used in the case of linear separability, with few parameters and fast speed, and the classification effect is relatively ideal for general data; the Gaussian kernel function is mainly used in the case of linear inseparability, with many parameters, and the classification results are too dependent on the parameters, requiring a large number of parameter attempts. Extract the distribution of some sample feature vectors, use the linear kernel function and the Gaussian kernel function for calculation respectively, the input of the model is the feature vectors Z1 and Z2, and set the following 4 groups of kernel functions for training.

[0116] The training results are shown in Table 1 below, which confirms that the overall classification effect of using the Gaussian kernel function is better than that of the linear kernel function.

[0117]

[0118]

[0119] Table 1

[0120] As shown in Table 1, the parameter setting penalty factor C = 5, γ = 4, the SVM algorithm using the Gaussian kernel function has the highest accuracy and the least number of support vectors, which means that the vector machine model training and prediction speed is the fastest.

[0121] In this embodiment, by evaluating the evaluation index, calculating the output result of the vector machine model, and adjusting the function of the vector machine model, the accuracy of the judgment result output by the vector machine model can be improved.

[0122] In one embodiment, the hyperplane equation in the vector machine model, as well as the corresponding objective function and constraint conditions (i.e., step D2) are obtained, and the following steps D21-D22 are performed:

[0123] Step D21: Obtain the hyperplane equation in the vector machine model, and determine the objective function and constraint conditions including the penalty factor and slack variable according to the sample feature vector.

[0124] According to the hyperplane equation in the vector machine model, hard interval and soft interval objective functions and constraints can be adopted. According to the distribution of sample feature vectors, it can be determined that the soft interval of objective functions and constraints with penalty factors and slack variables is more suitable for the vector machine model. Some points of sample feature vectors can be allowed to be wrong, making the vector machine model more generalizable.

[0125] Step D22: The objective function and constraint conditions including the penalty factor and the slack variable are used as the objective function and constraint conditions corresponding to the hyperplane equation.

[0126] It should be noted that after the training vector machine model is completed, multiple charging and discharging clips of power batteries are randomly obtained to verify the integrity of the car's charging and discharging process. For example, 9 groups of charging and discharging clips are randomly selected from all the charging and discharging clips, and 8 of them are used as training sets and 1 group is used as a test set in turn; at the same time, in order to reflect the integrity of the car's charging and discharging process, the clips are required to be more than 1000 frames, and the battery is selected as much as possible from full to empty, that is, the battery's state of charge (State of Charge, SOC) changes from small to large and then small. Random seed screening is used to select 9 groups of charging and discharging data. 9 groups of charging and discharging clips are obtained, and the detailed information of each clip is shown in Table 2 below. The data set is introduced in detail. It can be seen that the labels are relatively evenly distributed in each data segment of the entire life cycle, which is well representative.

[0127]

[0128] Table 2

[0129] In this embodiment, by determining the objective function and constraint conditions of the hyperplane in the vector machine model, the generalization of the trained vector machine model can be improved.

[0130] Figure 2 is a schematic flow chart of a safety warning method for a power battery according to another embodiment of the present application, such as Figure 2 As shown, the method comprises the following steps:

[0131] S201, obtaining actual vehicle operation data, and extracting operating condition data of the power battery under charging conditions from the operation data.

[0132] S202, dividing the operating condition data into segments to obtain target segments.

[0133] S203, extracting a voltage data matrix corresponding to the target segment.

[0134] S204, calculating a voltage change rate difference ratio index and a voltage consistency difference index according to the voltage data matrix.

[0135] S205 , normalizing the voltage change rate difference ratio index and the voltage consistency difference index to obtain corresponding feature vectors.

[0136] S206, inputting the feature vector into the trained support vector machine model for calculation, and outputting judgment data calculated based on the support vector machine.

[0137] S207, determining a discrimination result corresponding to the target segment according to the discrimination data, where the discrimination result includes whether the target segment is a fault segment or a normal segment.

[0138] S208, identifying the faulty power battery based on the discrimination result of the target segment, and then determining the faulty vehicle.

[0139] S209: If the faulty power battery corresponds to the faulty segment, a power battery safety warning is issued.

[0140] The specific process from S201 to S209 has been described in detail in the above embodiment and will not be repeated here.

[0141] By adopting the technical solution of the embodiment of the present application, the operating condition data of the power battery during charging is obtained, and the operating condition data is divided into segments to obtain voltage data corresponding to the target segment; based on the voltage data, the voltage parameter index of the target segment is calculated; the voltage parameter index includes: a voltage change rate difference ratio index and a voltage consistency difference index; the voltage parameter index is normalized to obtain a feature vector, and the feature vector includes: a voltage change rate feature vector and a voltage consistency vector; the feature vector is input into a pre-trained vector machine model, and judgment data is output, and the judgment data includes whether the target segment is a fault segment or a normal segment; based on the judgment data, a safety warning is given to the power battery. It can be seen that by obtaining the operating condition data of the power battery during the charging period and calculating the voltage parameter index corresponding to the target segment, the effective data safety index of the power battery can be obtained, and through the learning method of the support vector machine model, the voltage parameter index is input into the support vector machine model to output the judgment data. According to the judgment data, it is judged whether the target segment is a fault segment. The operating condition data of the power battery can be fully analyzed, and then the vehicle with the power battery with the faulty segment can be given a safety warning. There is no need to discover the safety problem of the vehicle when the vehicle is in use. The power battery fault can be accurately diagnosed and the faulty vehicle can be identified to achieve early warning.

[0142] In summary, specific embodiments of the present subject matter have been described. Other embodiments are within the scope of the appended claims. In some cases, the actions recorded in the claims can be performed in a different order and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing can be advantageous.

[0143] The above is a safety warning method for a power battery provided in an embodiment of the present application. Based on the same idea, an embodiment of the present application also provides a safety warning device for a power battery.

[0144] Figure 3 1 is a schematic diagram of a safety warning device for a power battery according to an embodiment of the present invention. Figure 3 As shown, the safety warning device for the power battery includes: an acquisition module 31 , a calculation module 32 , a processing module 33 , and an output module 34 .

[0145] An acquisition module 31 is used to acquire the operating condition data of the power battery during charging, and divide the operating condition data into segments to obtain voltage data corresponding to the target segment;

[0146] The calculation module 32 is used to calculate the voltage parameter index of the target segment according to the voltage data; the voltage parameter index includes: a voltage change rate difference ratio index and a voltage consistency difference index;

[0147] The processing module 33 is used to perform normalization processing on the voltage parameter index to obtain a characteristic vector, wherein the characteristic vector includes: a voltage change rate characteristic vector and a voltage consistency vector;

[0148] The output module 34 is used to input the feature vector into a pre-trained vector machine model and output judgment data, wherein the judgment data includes whether the target segment is a fault segment or a normal segment; and based on the judgment data, a safety warning is issued to the power battery.

[0149] In one embodiment, the computing module 32 is specifically configured to:

[0150] Determine, according to the voltage data, a voltage difference between a target single cell in the power battery in a target frame of a target segment and a next frame, and determine an average voltage value of each single cell corresponding to the target frame and an average voltage value of each single cell corresponding to the next frame; wherein the power battery includes a plurality of single cells;

[0151] Determine the average voltage difference corresponding to the target frame according to the average voltage value of each single cell corresponding to the target frame and the average voltage value of each single cell corresponding to the next frame;

[0152] Based on the voltage difference and the average voltage difference, a voltage change rate difference ratio of a target single cell of the target segment is determined, and the voltage change rate difference ratio of the target single cell is used as a voltage change rate difference ratio indicator of the target segment.

[0153] In one embodiment, the calculation module 32 is further configured to:

[0154] Determine, according to the voltage data, the voltage average value of the target frame in the target segment and the voltage value of each single cell corresponding to the target frame in the target frame;

[0155] Based on the difference between the voltage value of each single cell corresponding to the target frame in the target frame and the voltage average value, a voltage consistency difference index of the target segment is determined.

[0156] In one embodiment, the processing module 33 is specifically configured to normalize the voltage change rate difference ratio index and the voltage consistency difference index based on a Min-Max normalization algorithm to obtain a corresponding feature vector.

[0157] In one embodiment, the device further includes a training module of a vector machine model, specifically including:

[0158] A first acquisition unit is used to acquire sample feature vectors of multiple sample power batteries when they are charged and corresponding sample fault fragments, and input the sample feature vectors into a vector machine model;

[0159] A second acquisition unit is used to acquire a hyperplane equation in a vector machine model, as well as a corresponding objective function and constraint conditions;

[0160] A mapping unit, used for mapping the sample feature vector to a high-dimensional space according to the kernel function in the vector machine model;

[0161] A processing unit, used to classify the sample feature vectors through a hyperplane equation and a kernel function, and output sample judgment data;

[0162] The training unit is used to adjust the kernel function and train the vector machine model based on the sample judgment data and the sample fault fragments.

[0163] In one embodiment, the training unit is specifically configured to:

[0164] Calculate evaluation indicators of the vector machine model according to the sample judgment data and the sample fault fragments, where the evaluation indicators include one or more of the accuracy, precision, recall rate and harmonic mean of the recall rate and precision rate of the sample judgment data;

[0165] According to the evaluation index, the kernel function is adjusted to a Gaussian kernel function.

[0166] In one embodiment, the second acquisition unit is specifically configured to:

[0167] Obtain the hyperplane equation in the vector machine model, and determine the objective function and constraint conditions including the penalty factor and slack variable according to the sample feature vector;

[0168] The objective function and constraints including the penalty factor and the slack variable are used as the objective function and constraints corresponding to the hyperplane equation.

[0169] By adopting the technical solution of the embodiment of the present application, the operating condition data of the power battery during charging is obtained, and the operating condition data is divided into segments to obtain voltage data corresponding to the target segment; based on the voltage data, the voltage parameter index of the target segment is calculated; the voltage parameter index includes: a voltage change rate difference ratio index and a voltage consistency difference index; the voltage parameter index is normalized to obtain a feature vector, and the feature vector includes: a voltage change rate feature vector and a voltage consistency vector; the feature vector is input into a pre-trained vector machine model, and judgment data is output, and the judgment data includes whether the target segment is a fault segment or a normal segment; based on the judgment data, a safety warning is given to the power battery. It can be seen that by obtaining the operating condition data of the power battery during the charging period and calculating the voltage parameter index corresponding to the target segment, the effective data safety index of the power battery can be obtained, and through the learning method of the support vector machine model, the voltage parameter index is input into the support vector machine model to output the judgment data. According to the judgment data, it is judged whether the target segment is a fault segment. The operating condition data of the power battery can be fully analyzed, and then the vehicle with the power battery with the faulty segment can be given a safety warning. There is no need to discover the safety problem of the vehicle when the vehicle is in use. The power battery fault can be accurately diagnosed and the faulty vehicle can be identified to achieve early warning.

[0170] Those skilled in the art should understand that Figure 3 The safety warning device for the power battery in the present invention can be used to implement the safety warning method for the power battery described above, and the detailed description thereof should be similar to that described in the method part above. To avoid tediousness, it will not be further described here.

[0171] Based on the same technical concept, an embodiment of the present application further provides an electronic device, which is used to execute the above-mentioned power battery safety warning method. Figure 4 A schematic diagram of the structure of an electronic device for implementing various embodiments of the present application. The electronic device may have relatively large differences due to different configurations or performances, and may include a processor (processor) 410, a communication interface (Communications Interface) 420, a memory (memory) 930 and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other through the communication bus 440. The processor 410 may call a computer program stored in the memory 430 and executable on the processor 410 to perform the following steps:

[0172] Acquire the operating condition data of the power battery during charging, and divide the operating condition data into segments to obtain voltage data corresponding to the target segment;

[0173] Calculate the voltage parameter index of the target segment according to the voltage data; the voltage parameter index includes: voltage change rate difference ratio index and voltage consistency difference index;

[0174] The voltage parameter index is normalized to obtain a characteristic vector, which includes: a voltage change rate characteristic vector and a voltage consistency vector;

[0175] The feature vector is input into a pre-trained vector machine model, and judgment data is output, wherein the judgment data includes whether the target segment is a faulty segment or a normal segment; and a safety warning is issued to the power battery based on the judgment data.

[0176] By adopting the technical solution of the embodiment of the present application, the operating condition data of the power battery during charging is obtained, and the operating condition data is divided into segments to obtain voltage data corresponding to the target segment; based on the voltage data, the voltage parameter index of the target segment is calculated; the voltage parameter index includes: a voltage change rate difference ratio index and a voltage consistency difference index; the voltage parameter index is normalized to obtain a feature vector, and the feature vector includes: a voltage change rate feature vector and a voltage consistency vector; the feature vector is input into a pre-trained vector machine model, and judgment data is output, and the judgment data includes whether the target segment is a fault segment or a normal segment; based on the judgment data, a safety warning is given to the power battery. It can be seen that by obtaining the operating condition data of the power battery during the charging period and calculating the voltage parameter index corresponding to the target segment, the effective data safety index of the power battery can be obtained, and through the learning method of the support vector machine model, the voltage parameter index is input into the support vector machine model to output the judgment data. According to the judgment data, it is judged whether the target segment is a fault segment. The operating condition data of the power battery can be fully analyzed, and then the vehicle with the power battery with the faulty segment can be given a safety warning. There is no need to discover the safety problem of the vehicle when the vehicle is in use. The power battery fault can be accurately diagnosed and the faulty vehicle can be identified to achieve early warning.

[0177] The specific execution steps can refer to the various steps of the above-mentioned power battery safety warning method embodiment, and can achieve the same technical effect. To avoid repetition, they will not be repeated here.

[0178] It should be noted that the electronic devices in the embodiments of the present application include: servers, terminals, or other devices except terminals.

[0179] The above electronic device structure does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. For example, the input unit may include a graphics processing unit (GPU) and a microphone, and the display unit may be configured with a display panel in the form of a liquid crystal display, an organic light-emitting diode, etc. The user input unit includes a touch panel and at least one of other input devices. The touch panel is also called a touch screen. Other input devices may include, but are not limited to, a physical keyboard, function keys (such as volume control buttons, switch buttons, etc.), a trackball, a mouse, and a joystick, which will not be repeated here.

[0180] The memory can be used to store software programs and various data. The memory may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instructions required for at least one function (such as a sound playback function, an image playback function, etc.), etc. In addition, the memory may include a volatile memory or a non-volatile memory, or the memory may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM) and direct memory bus random access memory (DRRAM).

[0181] The processor may include one or more processing units; optionally, the processor integrates an application processor and a modem processor, wherein the application processor mainly processes operations related to the operating system, user interface, and application programs, and the modem processor mainly processes wireless communication signals, such as a baseband processor. It is understandable that the modem processor may not be integrated into the processor.

[0182] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, each process of the above-mentioned method embodiment based on power battery safety warning is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0183] The processor is a processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0184] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned method embodiment based on power battery safety warning, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0185] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0186] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0187] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0188] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A safety early warning method for a power battery, characterized in that: The method comprises: Acquire operating condition data of the power battery during charging, and divide the operating condition data into segments to obtain voltage data corresponding to the target segment; According to the voltage data, a voltage parameter index of the target segment is calculated; the voltage parameter index includes: a voltage change rate difference ratio index and a voltage consistency difference index; wherein, according to the voltage data, a voltage difference between a target single cell in the power battery in a target frame and a next frame of the target segment is determined, and an average voltage value of each single cell corresponding to the target frame and an average voltage value of each single cell corresponding to the next frame are determined; wherein the power battery includes a plurality of single cells; according to the average voltage value of each single cell corresponding to the target frame and the average voltage value of each single cell corresponding to the next frame, an average voltage difference corresponding to the target frame is determined; based on the voltage difference and the average voltage difference, a voltage change rate difference ratio of the target single cell in the target segment is determined, and the voltage change rate difference ratio of the target single cell is used as a voltage change rate difference ratio index of the target segment; according to the voltage data, an average voltage value of the target frame in the target segment and a voltage value of each single cell corresponding to the target frame in the target frame are determined; based on the difference between the voltage value of each single cell corresponding to the target frame in the target frame and the voltage average value, a voltage consistency difference index of the target segment is determined; Normalizing the voltage parameter index to obtain a characteristic vector, wherein the characteristic vector includes: a voltage change rate characteristic vector and a voltage consistency vector; The feature vector is input into a pre-trained vector machine model, and judgment data is output, wherein the judgment data includes whether the target segment is a faulty segment or a normal segment. The vector machine model is a pre-trained vector machine model of a support vector machine algorithm, and is used to classify the feature vector and identify whether the target segment corresponding to the feature vector is faulty; based on the judgment data, a safety warning is issued for the power battery.

2. The method according to claim 1, characterized in that The voltage parameter index is normalized to obtain a characteristic vector, including: Based on the Min-Max normalization algorithm, the voltage change rate difference ratio index and the voltage consistency difference index are normalized to obtain corresponding feature vectors.

3. The method according to claim 1, characterized in that The training of the vector machine model includes: Acquire sample feature vectors of multiple sample power batteries when they are charged and corresponding sample fault fragments, and input the sample feature vectors into the vector machine model; Obtaining a hyperplane equation in the vector machine model, as well as corresponding objective functions and constraints; According to the kernel function in the vector machine model, mapping the sample feature vector to a high-dimensional space; Classify the sample feature vector using the hyperplane equation and the kernel function, and output sample judgment data; Based on the sample judgment data and the sample fault fragments, the kernel function is adjusted and the vector machine model is trained.

4. The method according to claim 3, characterized in that: The adjusting the kernel function based on the sample judgment data and the sample fault fragment comprises: Calculate the evaluation index of the vector machine model according to the sample judgment data and the sample fault fragment, wherein the evaluation index includes one or more of the accuracy, precision, recall rate and the harmonic mean of the recall rate and the precision rate of the sample judgment data; According to the evaluation index, the kernel function is adjusted to a Gaussian kernel function.

5. The method according to claim 3, characterized in that: The step of obtaining the hyperplane equation in the vector machine model, and the corresponding objective function and constraint conditions includes: Obtaining a hyperplane equation in the vector machine model, and determining an objective function and constraint conditions including a penalty factor and a slack variable according to the sample feature vector; The objective function and constraint conditions including the penalty factor and the slack variable are used as the objective function and constraint conditions corresponding to the hyperplane equation.

6. A safety warning device for a power battery, used to execute a safety warning method for a power battery according to any one of claims 1 to 5, characterized in that: The device comprises: An acquisition module is used to acquire the operating condition data of the power battery during charging, and divide the operating condition data into segments to obtain voltage data corresponding to the target segment; A calculation module is used to calculate the voltage parameter index of the target segment according to the voltage data; the voltage parameter index includes: a voltage change rate difference ratio index and a voltage consistency difference index; wherein, according to the voltage data, the voltage difference between the target single cell in the power battery in the target frame and the next frame of the target segment is determined, and the average voltage value of each single cell corresponding to the target frame and the average voltage value of each single cell corresponding to the next frame are determined; wherein the power battery includes a plurality of single cells; according to the average voltage value of each single cell corresponding to the target frame and the average voltage value of each single cell corresponding to the next frame value, determine the average voltage difference value corresponding to the target frame; determine the voltage change rate difference ratio of the target single battery in the target segment based on the voltage difference value and the average voltage difference value, and use the voltage change rate difference ratio of the target single battery as the voltage change rate difference ratio index of the target segment; determine the voltage average value of the target frame in the target segment and the voltage value of each single battery corresponding to the target frame in the target frame according to the voltage data; determine the voltage consistency difference index of the target segment based on the difference between the voltage value of each single battery corresponding to the target frame in the target frame and the voltage average value; A processing module, used for normalizing the voltage parameter index to obtain a characteristic vector, wherein the characteristic vector includes: a voltage change rate characteristic vector and a voltage consistency vector; An output module is used to input the feature vector into a pre-trained vector machine model and output judgment data, wherein the judgment data includes whether the target segment is a faulty segment or a normal segment. The vector machine model is a pre-trained vector machine model of a support vector machine algorithm, and is used to classify the feature vector and identify whether the target segment corresponding to the feature vector is faulty; based on the judgment data, a safety warning is issued for the power battery.

7. An electronic device, characterized in that: It comprises a processor and a memory electrically connected to the processor, the memory stores a computer program, and the processor is used to call and execute the computer program from the memory to implement a safety warning method for a power battery as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The storage medium is used to store a computer program, and the computer program can be executed by a processor to implement a safety warning method for a power battery as described in any one of claims 1 to 5.

Images