Power battery safety early warning method, device, equipment, medium and program product

Abstract

The invention provides a power battery safety early warning method, device and equipment, a medium and a program product, and relates to the technical field of new energy automobiles. The power battery safety early warning method comprises the following steps: according to a power battery safety early warning model and current charging behavior data of a target power battery, obtaining a safety early warning result used for indicating whether the target power battery has an abnormal risk; wherein the power safety early warning model is obtained by training according to a charging behavior synthesis data set and a first historical charging behavior data set of a plurality of power batteries; the charging behavior synthesis data set is obtained by performing data synthesis processing according to a second historical charging behavior data set of the power battery with the abnormal risk in the plurality of power batteries. According to the scheme, the abnormal risk early warning of the power battery is carried out in combination with the charging behavior of the user, and the prediction accuracy of the abnormal risk of the power battery is obviously improved.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy vehicles, and in particular to a power battery safety early warning method, device, equipment, medium and program product. Background Art

[0002] With the increase in the number of new energy vehicles and the extension of battery service time, new energy vehicle fires have become more and more frequent in recent years. The occurrence of thermal runaway of new energy vehicles not only causes huge economic losses and affects user experience, but also has a negative impact on the brand power and reputation of manufacturers.

[0003] In the existing technology, power battery safety early warning technology is mostly based on the internal mechanism of the battery cell, such as early warning of abnormal risks of the power battery by analyzing the voltage difference and internal resistance of the battery. However, the abnormal risk early warning of the power battery based on the internal mechanism of the battery cell has the problem of poor expected effect and low prediction accuracy. Summary of the invention

[0004] The technical solution of the present invention aims to provide a power battery safety warning method, device, equipment, medium and program product, which are used to solve the problem of low accuracy of abnormal risk warning of power batteries in the prior art.

[0005] To achieve the above object, the present invention is achieved by:

[0006] In a first aspect, an embodiment of the present invention provides a power battery safety early warning method, comprising:

[0007] According to the power battery safety warning model and the current charging behavior data of the target power battery, obtaining a safety warning result indicating whether the target power battery has an abnormal risk;

[0008] Among them, the power safety warning model is obtained by training based on a charging behavior synthetic data set and a first historical charging behavior data set of multiple power batteries; the charging behavior synthetic data set is obtained by data synthesis processing based on a second historical charging behavior data set of power batteries with abnormal risks among the multiple power batteries.

[0009] Optionally, the power battery safety early warning method further comprises:

[0010] Follow the steps below to obtain the power battery safety warning model:

[0011] Acquire a first historical charging behavior data set of the plurality of power batteries; wherein the number of power batteries with abnormal risks among the plurality of power batteries is less than the number of power batteries without abnormal risks;

[0012] Acquire, according to the first historical charging behavior data set, a second historical charging behavior data set of a power battery having an abnormal risk among the plurality of power batteries;

[0013] Performing data synthesis processing on the second historical charging behavior data set to obtain the charging behavior synthetic data set;

[0014] Obtaining a training data set according to the charging behavior synthetic data set and the first historical charging behavior data set;

[0015] The preset classification model is trained according to the training data set to obtain the power battery safety warning model.

[0016] Optionally, the power battery safety early warning method, wherein, according to the first historical charging behavior data set, obtaining a second historical charging behavior data set of a power battery with abnormal risk among the plurality of power batteries comprises:

[0017] Performing data cleaning on the first historical charging behavior data set to remove missing data and abnormal data, and obtaining the cleaned first historical charging behavior data set;

[0018] Normalizing the first historical charging behavior data set after cleaning to obtain a normalized first historical charging behavior data set;

[0019] Performing data statistical analysis on the normalized first historical charging behavior data set to obtain the first historical charging behavior data set after statistical analysis;

[0020] A second historical charging behavior data set of a power battery having an abnormal risk among the plurality of power batteries is obtained according to the first historical charging behavior data set after statistical analysis.

[0021] Optionally, in the power battery safety early warning method, performing data statistical analysis on the normalized first historical charging behavior data set to obtain the first historical charging behavior data set after statistical analysis includes:

[0022] performing a correlation analysis on the normalized first historical charging behavior data set, removing historical charging behavior data having a Pearson correlation coefficient equal to a preset value in the normalized first historical charging behavior data set, and obtaining the first historical charging behavior data set after the correlation analysis;

[0023] The historical charging behavior data whose variance is smaller than a threshold value in the first historical charging behavior data set after correlation analysis is eliminated to obtain the first historical charging behavior data set after statistical analysis.

[0024] Optionally, the power battery safety warning method, wherein performing data synthesis processing according to the second historical charging behavior data set to obtain the charging behavior synthetic data set, includes:

[0025] For each historical charging behavior data in the second historical charging behavior data set, obtaining at least one target data corresponding to the historical charging behavior data in the second historical charging behavior data set;

[0026] Obtaining an absolute value of a difference between the historical charging behavior data and the target data;

[0027] According to the product of the absolute value and the random number and the sum of the historical charging behavior data, one of the charging behavior synthetic data in the charging behavior synthetic data set is obtained.

[0028] Optionally, in the power battery safety warning method, obtaining at least one target data corresponding to the historical charging behavior data in the second historical charging behavior data set includes:

[0029] Obtaining the Euclidean distance between the historical charging behavior data and target historical charging behavior data; wherein the target historical charging behavior data is historical charging behavior data in the second historical charging behavior data set that belongs to the same dimension as the historical charging behavior data;

[0030] According to the Euclidean distance, obtaining a plurality of neighboring data of the historical charging behavior data;

[0031] The at least one target data is selected from the plurality of neighboring data according to a preset sampling ratio.

[0032] In a second aspect, an embodiment of the present invention provides a power battery safety warning device, comprising:

[0033] A first acquisition module, configured to acquire, according to a power battery safety warning model and current charging behavior data of a target power battery, a safety warning result indicating whether the target power battery has an abnormal risk;

[0034] Among them, the power safety warning model is obtained by training based on a charging behavior synthetic data set and a first historical charging behavior data set of multiple power batteries; the charging behavior synthetic data set is obtained by data synthesis processing based on a second historical charging behavior data set of power batteries with abnormal risks among the multiple power batteries.

[0035] In a third aspect, an embodiment of the present invention provides a power battery safety warning device, comprising a processor, a memory, and a program or instruction stored in the memory and executable on the processor; when the processor executes the program or instruction, the power battery safety warning method as described in the first aspect is implemented.

[0036] In a fourth aspect, an embodiment of the present invention provides a readable storage medium, on which a program is stored, and when the program is executed by a processor, the power battery safety warning method as described in the first aspect is implemented.

[0037] In a fifth aspect, an embodiment of the present invention provides a computer program product, including computer instructions, which, when executed by a processor, implement the power battery safety warning method as described in the first aspect.

[0038] The above technical solution of the present invention has at least the following beneficial effects:

[0039] The power battery safety warning method described in the embodiment of the present invention is adopted to obtain a safety warning result indicating whether the target power battery has an abnormal risk according to the power battery safety warning model and the current charging behavior data of the target power battery; wherein the power safety warning model is obtained by training based on a charging behavior synthetic data set and a first historical charging behavior data set of multiple power batteries; the charging behavior synthetic data set is obtained by data synthesis processing based on a second historical charging behavior data set of a power battery with an abnormal risk among the multiple power batteries. In this way, the power battery safety warning model is trained with the user's charging behavior data, and whether the power battery has an abnormal risk is predicted based on the power battery safety warning model, thereby effectively improving the prediction accuracy of the abnormal risk of the power battery. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic diagram of the process of the power battery safety early warning method according to an embodiment of the present invention;

[0041] Figure 2 Schematic diagram of the principle of a support vector machine in an embodiment of the present invention;

[0042] Figure 3 It is a schematic diagram of the structure of the power battery safety warning device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0044] Embodiments of the present invention address the problem of low accuracy in abnormal risk warning of power batteries in the prior art and provide a power battery safety warning method, device, equipment, medium and program product.

[0045] like Figure 1 As shown, an embodiment of the present invention provides a power battery safety early warning method, comprising:

[0046] S101, obtaining a safety warning result indicating whether the target power battery has an abnormal risk according to a power battery safety warning model and current charging behavior data of a target power battery;

[0047] Among them, the power safety warning model is obtained by training based on a charging behavior synthetic data set and a first historical charging behavior data set of multiple power batteries; the charging behavior synthetic data set is obtained by data synthesis processing based on a second historical charging behavior data set of power batteries with abnormal risks among the multiple power batteries.

[0048] It is understandable that the charging behavior data is data related to the user's charging behavior, and can be charging behavior data in multiple dimensions, including current dimension, temperature dimension, and state of charge dimension, such as current, single cell maximum voltage, single cell minimum voltage, single cell maximum temperature, single cell minimum temperature, state of charge, etc. It is also understandable that the charging behavior data also includes fast charging and slow charging indications, vehicle identification code, accumulated mileage, etc.

[0049] It should be noted that the embodiments of the present invention fully consider the user's charging behavior when issuing warnings for abnormal risks of the power battery, and associate the user's charging behavior with the abnormal risks of the power battery. Compared with the prior art, the abnormal risk warning of the power battery is mainly based on the internal mechanism of the battery cell, and the prediction accuracy is greatly improved.

[0050] As shown in Table 1 below, the current charging behavior data of the target power battery includes the proportion of fast charging times, the average state of charge at the end of charging, and the proportion of time when charging is greater than 40 degrees. The current charging behavior data of the target power is input into the power battery safety warning model, and the safety warning result is output, where a safety warning result of 1 indicates that the power battery has an abnormal risk, and a safety warning result of 0 indicates that the power battery has no abnormal risk.

[0051]

[0052] Table 1: Examples of safety warning results

[0053] In one embodiment, optionally, the method further comprises:

[0054] Follow the steps below to obtain the power battery safety warning model:

[0055] Acquire a first historical charging behavior data set of the plurality of power batteries; wherein the number of power batteries with abnormal risks among the plurality of power batteries is less than the number of power batteries without abnormal risks;

[0056] Acquire, according to the first historical charging behavior data set, a second historical charging behavior data set of a power battery having an abnormal risk among the plurality of power batteries;

[0057] Performing data synthesis processing on the second historical charging behavior data set to obtain the charging behavior synthetic data set;

[0058] Obtaining a training data set according to the charging behavior synthetic data set and the first historical charging behavior data set;

[0059] The preset classification model is trained according to the training data set to obtain the power battery safety warning model.

[0060] In an embodiment of the present invention, a first historical charging behavior data set of power batteries of multiple vehicles within a preset time period (such as nearly one year) before the current time can be obtained, which includes a historical charging behavior data set of power batteries with abnormal risks (i.e., a second historical charging behavior data set), and of course also includes a historical charging behavior data set of power batteries without abnormal risks. It can be understood that the number of power batteries with abnormal risks is less than the number of power batteries without abnormal risks.

[0061] Perform data synthesis processing on the second historical charging behavior data set to obtain a charging behavior synthetic data set, and use the charging behavior synthetic data set and the first historical charging behavior data set to form a sample data set; use the random selection principle and divide the sample data set into a training data set, a test data set, and a verification data set according to a preset ratio. Taking the preset ratio of 7:2:1 as an example, 70% of the data randomly selected from the sample data set is used as a training data set, 20% of the data is used as a test data set, and 10% of the data is used as a verification data set.

[0062] Based on the training data set, a power battery safety warning model is built using the preset classification model.

[0063] It should be noted that the preset classification model is a support vector machine (SVM). The basic idea of ​​SVM learning is to find a separating hyperplane that can correctly divide the training data set and has the largest geometric interval. Figure 2 As shown, w·x+b=0 is the separating hyperplane. For a linearly separable data set, there are infinitely many such hyperplanes (i.e., perceptrons), but the separating hyperplane with the largest geometric interval is unique.

[0064] It should also be noted that, optionally, the method further includes:

[0065] Obtaining a test data set according to the charging behavior synthetic data set and the first historical charging behavior data set;

[0066] Using the test data set and the power battery safety early warning model, a confusion matrix is ​​constructed;

[0067] According to the confusion matrix, the power battery safety warning model is evaluated to obtain a model evaluation result.

[0068] The test data set has been described in the above embodiment and will not be described again here.

[0069] The test data set is input into the power battery safety warning model to obtain a safety warning result (i.e., a predicted value), and a confusion matrix is ​​constructed according to the true value and the predicted value corresponding to the test data set, as shown in Table 2. The symbols in Table 2 are explained below:

[0070] Category 1 indicates a safety warning result that the power battery has an abnormal risk;

[0071] Category 2 indicates that there is no safety warning result of abnormal risk in the power battery;

[0072] a represents the number of safety warning results whose predicted values ​​are the safety warning results of abnormal risk of power battery and whose true values ​​are the safety warning results of abnormal risk of power battery, that is, the number of predicted values ​​that are consistent with the true values ​​(that is, the prediction is correct);

[0073] b represents the number of safety warning results where the predicted value is that there is no abnormal risk in the power battery and the true value is that there is an abnormal risk in the power battery, that is, the number of predicted values ​​that do not match the true value (that is, the prediction is incorrect);

[0074] c represents the number of safety warning results where the predicted value is that the power battery has an abnormal risk and the true value is that the power battery does not have an abnormal risk, that is, the number of predicted values ​​that do not match the true value (that is, the prediction is incorrect);

[0075] d indicates that the predicted value is the safety warning result that there is no abnormal risk in the power battery, and the true value is the number of safety warning results that there is no abnormal risk in the power battery, that is, the number of predicted values ​​that are consistent with the true values ​​(that is, the prediction is correct).

[0076]

[0077] Table 2: Confusion matrix example

[0078] Furthermore, the power battery safety warning model is evaluated using the confusion matrix to obtain a model evaluation result indicating a prediction effect of the power battery safety warning model.

[0079] The power battery safety warning model is evaluated by at least one of the following evaluation methods:

[0080] Accuracy evaluation method:

[0081] The number of correct predictions divided by the total number of test data sets equals the model prediction accuracy, accuracy = (a+d) / (a+b+c+d). It can be understood that the closer the accuracy is to 1, the better the model.

[0082] Accuracy evaluation method:

[0083] Also known as the precision rate, it indicates the proportion of the true minority class among all the samples predicted to be the minority class. Precision = a / (a+c). In SVM, the higher the precision, the more minority classes are captured and the more accurate the prediction of the minority class. Conversely, the lower the precision, the more majority classes are mistakenly captured. Precision is a measure of the cost of misjudging the majority class.

[0084] Recall evaluation method:

[0085] Also known as sensitivity, true positive rate, and recall rate, it represents the proportion of samples that are predicted correctly among all samples that are actually class 1. Recall rate = a / (a+b).

[0086] It should be noted that the numerators of recall and precision are the same, but the denominators are different. Recall and precision are trade-offs, and the balance between the two represents the balance between the need to capture the minority class and the need to avoid accidentally harming the majority class. Which side to lean towards depends on business needs.

[0087] In one embodiment, optionally, obtaining a second historical charging behavior data set of a power battery with abnormal risk among the plurality of power batteries according to the first historical charging behavior data set includes:

[0088] Performing data cleaning on the first historical charging behavior data set to remove missing data and abnormal data, and obtaining the cleaned first historical charging behavior data set;

[0089] Normalizing the first historical charging behavior data set after cleaning to obtain a normalized first historical charging behavior data set;

[0090] Performing data statistical analysis on the normalized first historical charging behavior data set to obtain the first historical charging behavior data set after statistical analysis;

[0091] A second historical charging behavior data set of a power battery having an abnormal risk among the plurality of power batteries is obtained according to the first historical charging behavior data set after statistical analysis.

[0092] In the embodiment of the present invention, firstly, data cleaning is performed on the first historical charging behavior data set to remove missing data and abnormal data, and the cleaned first historical charging behavior data set is obtained. Specifically, the data cleaning operation is as follows:

[0093] For missing data, if the missing data accounts for less than 20%, the data is directly filled by taking the mean value; if the missing data accounts for more than 20% and less than 50%, the missing data and non-missing data are marked; if the missing data accounts for more than 50%, the data is deleted.

[0094] Abnormal data whose values ​​exceed the threshold range need to be deleted directly.

[0095] Then, the first historical charging behavior data set after cleaning is normalized so that all data are at the same dimension level to obtain the normalized first historical charging behavior data set. Specifically, the data normalization operation is as follows:

[0096]

[0097] Wherein, x represents any data in the first historical charging behavior data set; ′ Represents the corresponding normalized data; min represents the minimum value of the data; max represents the maximum value of the data.

[0098] Furthermore, a data statistical analysis is performed on the normalized first historical charging behavior data set to obtain the first historical charging behavior data set after statistical analysis, so that a second historical charging behavior data set of the power battery with abnormal risk among the multiple power batteries can be obtained from the first historical charging behavior data set. It is understandable that the first historical charging behavior data set includes the second historical charging behavior data set.

[0099] In one implementation manner, optionally, performing data statistical analysis on the normalized first historical charging behavior data set to obtain the first historical charging behavior data set after statistical analysis includes:

[0100] performing a correlation analysis on the normalized first historical charging behavior data set, removing historical charging behavior data having a Pearson correlation coefficient equal to a preset value in the normalized first historical charging behavior data set, and obtaining the first historical charging behavior data set after the correlation analysis;

[0101] The historical charging behavior data whose variance is smaller than a threshold value in the first historical charging behavior data set after correlation analysis is eliminated to obtain the first historical charging behavior data set after statistical analysis.

[0102] In an embodiment of the present invention, a correlation analysis is performed on the normalized first historical charging behavior data set, the Pearson correlation coefficient between any two historical charging behavior data in the first historical charging behavior data set is calculated, and the two historical charging behavior data whose Pearson correlation coefficient is equal to a preset value are eliminated. The value range of the Pearson correlation coefficient is between -1 and 1, -1 indicates that the two historical charging behavior data are completely negatively correlated, 1 indicates that the two historical charging behavior data are completely positively correlated, and 0 indicates that the two historical charging behavior data are completely unrelated. The closer to 0, the weaker the correlation. Here, the preset value can be a number greater than 0.8 or less than -0.8.

[0103] The Pearson correlation coefficient is shown in the following formula:

[0104]

[0105] Where r represents the Pearson correlation coefficient; x i and i Respectively represent two historical charging behavior data.

[0106] Furthermore, historical charging behavior data having a variance less than a threshold value in the first historical charging behavior data set after correlation analysis is calculated, thereby obtaining the first historical charging behavior data set after statistical analysis.

[0107] In one implementation manner, optionally, performing data synthesis processing according to the second historical charging behavior data set to obtain the charging behavior synthetic data set includes:

[0108] For each historical charging behavior data in the second historical charging behavior data set, obtaining at least one target data corresponding to the historical charging behavior data in the second historical charging behavior data set;

[0109] Obtaining an absolute value of a difference between the historical charging behavior data and the target data;

[0110] According to the product of the absolute value and the random number and the sum of the historical charging behavior data, one of the charging behavior synthetic data in the charging behavior synthetic data set is obtained.

[0111] In one implementation manner, optionally, obtaining at least one target data corresponding to the historical charging behavior data in the second historical charging behavior data set includes:

[0112] Obtaining the Euclidean distance between the historical charging behavior data and target historical charging behavior data; wherein the target historical charging behavior data is historical charging behavior data in the second historical charging behavior data set that belongs to the same dimension as the historical charging behavior data;

[0113] According to the Euclidean distance, obtaining a plurality of neighboring data of the historical charging behavior data;

[0114] The at least one target data is selected from the plurality of neighboring data according to a preset sampling ratio.

[0115] In an embodiment of the present invention, since in the first historical charging behavior data set, the historical charging behavior data set of power batteries with abnormal risks is less than the historical charging behavior data set of power batteries without abnormal risks, there is a problem of unbalanced training data set samples. In order to solve this problem, it is necessary to adopt the synthetic minority oversampling technique (Synthetic Minority Oversampling Technique, SMOTE). SMOTE is an improved scheme based on the random oversampling algorithm. Since random oversampling adopts a simple sample duplication strategy to increase minority class samples, this easily leads to the problem of model overfitting, that is, the information learned by the model is too specific and not generalized enough. The basic idea of ​​SMOTE is to analyze the minority class samples and artificially synthesize new samples based on the minority class samples and add them to the data set.

[0116] Specifically, for each historical charging behavior data in the second historical charging behavior data set, the Euclidean distance between the historical charging behavior data and the target historical charging behavior data is obtained, and multiple neighboring data of the historical charging behavior data are obtained according to the Euclidean distance. According to the sample imbalance ratio, a preset sampling multiplier is determined, and at least one target data corresponding to the historical charging behavior data is selected from the multiple neighboring data.

[0117] Then, for each of the target data, charging behavior synthetic data is synthesized with the corresponding historical charging behavior data according to the following formula.

[0118] xnew=x+rand(0,1)*|x-xn|

[0119] Among them, xnew represents the synthetic data of charging behavior; x represents the charging behavior data; and xn represents the target data corresponding to x.

[0120] It should be noted that the charging behavior data related to at least one of the current dimension, the temperature dimension and the state of charge dimension in the embodiment of the present invention is shown in Table 3 below.

[0121]

[0122] Table 3: Example of charging behavior data

[0123] In summary, the power battery safety warning method described in the embodiment of the present invention is adopted to introduce the user's charging behavior data into the power battery safety warning method, and the charging behavior data involves multiple dimensions, such as current, temperature, state of charge, etc., and a power battery safety warning model is constructed to effectively improve the prediction accuracy of abnormal risks of power batteries.

[0124] like Figure 3 As shown, an embodiment of the present invention further provides a power battery safety warning device, comprising:

[0125] A first acquisition module 301 is used to acquire a safety warning result indicating whether the target power battery has an abnormal risk according to the power battery safety warning model and the current charging behavior data of the target power battery;

[0126] Among them, the power safety warning model is obtained by training based on a charging behavior synthetic data set and a first historical charging behavior data set of multiple power batteries; the charging behavior synthetic data set is obtained by data synthesis processing based on a second historical charging behavior data set of power batteries with abnormal risks among the multiple power batteries.

[0127] Optionally, the power battery safety warning device further comprises:

[0128] The second acquisition module is used to acquire the power battery safety warning model through the following submodules:

[0129] A first acquisition submodule is used to acquire a first historical charging behavior data set of the plurality of power batteries; wherein the number of power batteries with abnormal risks among the plurality of power batteries is less than the number of power batteries without abnormal risks;

[0130] A second acquisition submodule, configured to acquire, according to the first historical charging behavior data set, a second historical charging behavior data set of a power battery having an abnormal risk among the plurality of power batteries;

[0131] A first obtaining submodule, configured to perform data synthesis processing according to the second historical charging behavior data set to obtain the charging behavior synthetic data set;

[0132] A second acquisition submodule, configured to obtain a training data set according to the charging behavior synthetic data set and the first historical charging behavior data set;

[0133] The third acquisition submodule is used to train the preset classification model according to the training data set to obtain the power battery safety warning model.

[0134] Optionally, in the power battery safety warning device, the second acquisition submodule includes:

[0135] a cleaning unit, configured to clean the first historical charging behavior data set to remove missing data and abnormal data, and obtain the cleaned first historical charging behavior data set;

[0136] a normalization unit, configured to perform data normalization on the first historical charging behavior data set after cleaning to obtain a normalized first historical charging behavior data set;

[0137] A statistical analysis unit, configured to perform a data statistical analysis on the normalized first historical charging behavior data set to obtain the first historical charging behavior data set after statistical analysis;

[0138] The first acquisition unit is configured to acquire a second historical charging behavior data set of a power battery having an abnormal risk among the plurality of power batteries according to the first historical charging behavior data set after statistical analysis.

[0139] Optionally, in the power battery safety early warning device, the statistical analysis unit is specifically used to:

[0140] performing a correlation analysis on the normalized first historical charging behavior data set, removing historical charging behavior data having a Pearson correlation coefficient equal to a preset value in the normalized first historical charging behavior data set, and obtaining the first historical charging behavior data set after the correlation analysis;

[0141] The historical charging behavior data whose variance is smaller than a threshold value in the first historical charging behavior data set after correlation analysis is eliminated to obtain the first historical charging behavior data set after statistical analysis.

[0142] Optionally, in the power battery safety warning device, the first obtaining submodule includes:

[0143] a second acquisition unit, configured to acquire, for each piece of historical charging behavior data in the second historical charging behavior data set, at least one target data corresponding to the historical charging behavior data in the second historical charging behavior data set;

[0144] A third acquisition unit, configured to acquire an absolute value of a difference between the historical charging behavior data and the target data;

[0145] The obtaining unit is used to obtain one of the charging behavior synthetic data in the charging behavior synthetic data set according to the product of the absolute value and the random number and the sum of the historical charging behavior data.

[0146] Optionally, in the power battery safety warning device, the second acquisition unit is specifically used to:

[0147] Obtaining the Euclidean distance between the historical charging behavior data and target historical charging behavior data; wherein the target historical charging behavior data is historical charging behavior data in the second historical charging behavior data set that belongs to the same dimension as the historical charging behavior data;

[0148] According to the Euclidean distance, obtaining a plurality of neighboring data of the historical charging behavior data;

[0149] The at least one target data is selected from the plurality of neighboring data according to a preset sampling ratio.

[0150] The power battery safety warning device provided in the embodiment of the present invention can execute the above-mentioned power battery safety warning method embodiment, and its implementation principle and technical effect are similar, which will not be repeated in this embodiment.

[0151] An embodiment of the present invention also provides a power battery safety warning device, including a processor, a memory, and a program or instruction stored in the memory and executable on the processor; when the processor executes the program or instruction, the power battery safety warning method as described above is implemented, which will not be repeated here.

[0152] A specific embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps in the above-mentioned power battery safety warning method are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0153] In addition, an embodiment of the present invention further provides a computer program product, including computer instructions, which, when executed by a processor, implement the above Figure 1 The various processes of the method embodiment shown can achieve the same technical effect, and will not be described again here to avoid repetition.

[0154] In the several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0155] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may be physically included separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0156] The above-mentioned integrated unit implemented in the form of a software functional unit can be stored in a computer-readable storage medium. The above-mentioned software functional unit is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform some steps of the sending and receiving methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store program codes.

[0157] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A power battery safety early warning method, characterized in that: include: According to the power battery safety warning model and the current charging behavior data of the target power battery, obtaining a safety warning result indicating whether the target power battery has an abnormal risk; Among them, the power safety warning model is obtained by training based on a charging behavior synthetic data set and a first historical charging behavior data set of multiple power batteries; the charging behavior synthetic data set is obtained by data synthesis processing based on a second historical charging behavior data set of power batteries with abnormal risks among the multiple power batteries.

2. The power battery safety early warning method according to claim 1, characterized in that: The method further comprises: Follow the steps below to obtain the power battery safety warning model: Acquire a first historical charging behavior data set of the plurality of power batteries; wherein the number of power batteries with abnormal risks among the plurality of power batteries is less than the number of power batteries without abnormal risks; Acquire, according to the first historical charging behavior data set, a second historical charging behavior data set of a power battery having an abnormal risk among the plurality of power batteries; Performing data synthesis processing on the second historical charging behavior data set to obtain the charging behavior synthetic data set; Obtaining a training data set according to the charging behavior synthetic data set and the first historical charging behavior data set; The preset classification model is trained according to the training data set to obtain the power battery safety warning model.

3. The power battery safety early warning method according to claim 2, characterized in that: Acquiring, according to the first historical charging behavior data set, a second historical charging behavior data set of a power battery having an abnormal risk among the plurality of power batteries, including: Performing data cleaning on the first historical charging behavior data set to remove missing data and abnormal data, and obtaining the cleaned first historical charging behavior data set; Normalizing the first historical charging behavior data set after cleaning to obtain a normalized first historical charging behavior data set; Performing data statistical analysis on the normalized first historical charging behavior data set to obtain the first historical charging behavior data set after statistical analysis; A second historical charging behavior data set of a power battery having an abnormal risk among the plurality of power batteries is obtained according to the first historical charging behavior data set after statistical analysis.

4. The power battery safety early warning method according to claim 3, characterized in that: Performing data statistical analysis on the normalized first historical charging behavior data set to obtain the first historical charging behavior data set after statistical analysis, including: performing a correlation analysis on the normalized first historical charging behavior data set, removing historical charging behavior data having a Pearson correlation coefficient equal to a preset value in the normalized first historical charging behavior data set, and obtaining the first historical charging behavior data set after the correlation analysis; The historical charging behavior data whose variance is smaller than a threshold value in the first historical charging behavior data set after correlation analysis is eliminated to obtain the first historical charging behavior data set after statistical analysis.

5. The power battery safety early warning method according to claim 3, characterized in that: Performing data synthesis processing according to the second historical charging behavior data set to obtain the charging behavior synthetic data set includes: For each historical charging behavior data in the second historical charging behavior data set, obtaining at least one target data corresponding to the historical charging behavior data in the second historical charging behavior data set; Obtaining an absolute value of a difference between the historical charging behavior data and the target data; According to the product of the absolute value and the random number and the sum of the historical charging behavior data, one of the charging behavior synthetic data in the charging behavior synthetic data set is obtained.

6. The power battery safety early warning method according to claim 5, characterized in that: Acquiring at least one target data corresponding to the historical charging behavior data in the second historical charging behavior data set includes: Obtaining the Euclidean distance between the historical charging behavior data and target historical charging behavior data; wherein the target historical charging behavior data is historical charging behavior data in the second historical charging behavior data set that belongs to the same dimension as the historical charging behavior data; According to the Euclidean distance, obtaining a plurality of neighboring data of the historical charging behavior data; The at least one target data is selected from the plurality of neighboring data according to a preset sampling ratio.

7. A power battery safety warning device, characterized in that: include: A first acquisition module, configured to acquire a safety warning result indicating whether the target power battery has an abnormal risk according to the power battery safety warning model and current charging behavior data of the target power battery; Among them, the power safety warning model is obtained by training based on a charging behavior synthetic data set and a first historical charging behavior data set of multiple power batteries; the charging behavior synthetic data set is obtained by data synthesis processing based on a second historical charging behavior data set of power batteries with abnormal risks among the multiple power batteries.

8. A power battery safety warning device, comprising a processor, a memory, and a program or instruction stored in the memory and executable on the processor; characterized in that: When the processor executes the program or instruction, the power battery safety warning method according to any one of claims 1 to 6 is implemented.

9. A readable storage medium, characterized in that: The readable storage medium stores a program, and when the program is executed by the processor, the power battery safety warning method according to any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that It comprises computer instructions, which, when executed by a processor, implement the power battery safety warning method as claimed in any one of claims 1 to 6.

Images