Method and device for constructing capacity retention rate estimation model and computer equipment

Abstract

The invention relates to a method and device for constructing a capacity retention rate estimation model and computer equipment. The method comprises the following steps: acquiring a plurality of feature data sets according to charging and discharging data of a plurality of sample batteries; each feature data set comprises a plurality of sample features; training the initial random forest model according to the plurality of feature data sets, and constructing to obtain a target prediction model for predicting the capacity retention ratio of any target battery; the loss of the initial random forest model in the training process is determined according to the feature weight of each sample feature. Feature weights are introduced into the loss of node splitting, so that the sensitivity of the model to important features is improved, the training efficiency of the initial random forest model is improved, and the accuracy of the target prediction model is improved; and the initial random forest model is trained according to the plurality of feature data sets, the nonlinear relationship between sample features is fully utilized, and the prediction precision of the target prediction model on the battery later-stage capacity retention rate is improved.

Description

Technical Field

[0001] The present application relates to the technical field of lithium-ion batteries, and in particular to a method, device and computer equipment for constructing a capacity retention rate estimation model. Background Art

[0002] Lithium-ion batteries have the characteristics of high energy density, long cycle life and environmental friendliness. They are widely used in portable electronic devices, electric vehicles and energy storage systems. However, as the use time of lithium-ion batteries increases, the capacity of lithium-ion batteries will gradually decay, leading to the termination of the battery. Therefore, estimating the capacity retention rate of lithium-ion batteries is of great significance for evaluating battery performance, guiding battery design, optimizing battery management, discovering potential problems and reducing development costs and time.

[0003] Typically, the capacity retention rate of lithium-ion batteries is estimated by collecting data over the entire life cycle of the lithium-ion batteries.

[0004] However, conventional techniques have a problem of low efficiency in estimating the capacity retention rate of lithium-ion batteries. Summary of the invention

[0005] Based on this, it is necessary to provide a method, device and computer equipment for constructing a capacity retention rate estimation model that can improve the estimation efficiency of the capacity retention rate of a battery in response to the above technical problems.

[0006] In a first aspect, the present application provides a method for constructing a capacity retention rate estimation model, comprising:

[0007] Acquire multiple feature data sets according to the charge and discharge data of multiple sample batteries; each of the feature data sets includes multiple sample features;

[0008] The initial random forest model is trained according to the multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery; the loss of the initial random forest model during the training process is determined according to the feature weight of each of the sample features.

[0009] In one embodiment, the initial random forest model is trained according to the multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery, including:

[0010] Training the initial random forest model according to the multiple feature data sets to obtain a prediction model corresponding to each feature data set;

[0011] According to the prediction models corresponding to the characteristic data sets and the model weights of the prediction models, a target prediction model for predicting the capacity retention rate of any target battery is constructed.

[0012] In one embodiment, the initial random forest model is trained according to the multiple feature data sets to obtain a prediction model corresponding to each feature data set, including:

[0013] Inputting the sample features into the initial random forest model to obtain an initial prediction result; the initial random forest model is composed of a plurality of leaf nodes and a root node;

[0014] Determining the node loss of each leaf node in the initial random forest model according to the feature weight of each sample feature and the initial prediction result;

[0015] The initial random forest model is trained according to the node losses to obtain the prediction model corresponding to each feature data set.

[0016] In one embodiment, determining the node loss of each leaf node in the initial random forest model according to the feature weight of each sample feature and the initial prediction result includes:

[0017] Determine, according to the feature weights of the sample features, a first weighted mean square error of each leaf node in the initial random forest model and a second weighted mean square error of a child node associated with each leaf node;

[0018] The node loss of each leaf node in the initial random forest model is determined according to the first weighted mean square error of each leaf node and the second weighted mean square error of each child node.

[0019] In one embodiment, the method further comprises:

[0020] Determining the correlation between each of the sample characteristics and the capacity retention rate;

[0021] According to the correlation between each of the sample characteristics and the capacity retention rate, the sample weight of each of the sample characteristics is determined.

[0022] In one embodiment, the method further comprises:

[0023] Acquiring charging and discharging data of the target battery, and determining a feature data set corresponding to the target battery according to the charging and discharging data;

[0024] The characteristic data set corresponding to the target battery is input into the target prediction model, and the capacity retention rate of the target battery is obtained as an output.

[0025] In a second aspect, the present application also provides a device for constructing a capacity retention rate estimation model, comprising:

[0026] A first acquisition module, configured to acquire a plurality of feature data sets according to charge and discharge data of a plurality of sample batteries; each of the feature data sets includes a plurality of sample features;

[0027] A construction module is used to train an initial random forest model according to the multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery; the loss of the initial random forest model during the training process is determined according to the feature weight of each of the sample features.

[0028] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0029] Acquire multiple feature data sets according to the charge and discharge data of multiple sample batteries; each of the feature data sets includes multiple sample features;

[0030] The initial random forest model is trained according to the multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery; the loss of the initial random forest model during the training process is determined according to the feature weight of each of the sample features.

[0031] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0032] Acquire multiple feature data sets according to the charge and discharge data of multiple sample batteries; each of the feature data sets includes multiple sample features;

[0033] The initial random forest model is trained according to the multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery; the loss of the initial random forest model during the training process is determined according to the feature weight of each of the sample features.

[0034] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0035] Acquire multiple feature data sets according to the charge and discharge data of multiple sample batteries; each of the feature data sets includes multiple sample features;

[0036] The initial random forest model is trained according to the multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery; the loss of the initial random forest model during the training process is determined according to the feature weight of each of the sample features.

[0037] The construction method, device and computer equipment of the above-mentioned capacity retention rate estimation model obtain multiple feature data sets based on the charge and discharge data of multiple sample batteries; each feature data set includes multiple sample features; the initial random forest model is trained based on the multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery; the loss of the initial random forest model during the training process is determined according to the feature weight of each sample feature. The feature weight is introduced into the loss of node splitting to improve the sensitivity of the model to important features, thereby improving the efficiency of training the initial random forest model and improving the accuracy of the target prediction model; and the initial random forest model is trained based on multiple feature data sets, making full use of the nonlinear relationship between the sample features, and improving the prediction accuracy of the target prediction model for the later capacity retention rate of the battery. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 FIG. 1 is an application environment diagram of a method for constructing a capacity retention rate estimation model in an embodiment;

[0040] Figure 2 is a flow chart of a method for constructing a capacity retention rate estimation model in one embodiment;

[0041] Figure 3 is a flow chart of a method for constructing a capacity retention rate estimation model in another embodiment;

[0042] Figure 4 A performance diagram of predictions made by each prediction model according to a training set in one embodiment;

[0043] Figure 5 A performance diagram of predictions made by each prediction model according to a test set in one embodiment;

[0044] Figure 6 is a flow chart of a method for constructing a capacity retention rate estimation model in another embodiment;

[0045] Figure 7 is a flow chart of a method for constructing a capacity retention rate estimation model in another embodiment;

[0046] Figure 8 is a flow chart of a method for constructing a capacity retention rate estimation model in another embodiment;

[0047] Fig. 9 is a flow chart of a method for constructing a capacity retention rate estimation model in another embodiment;

[0048] Fig.10 is a flow chart of a method for constructing a capacity retention rate estimation model in another embodiment;

[0049] Fig.11 A structural block diagram of a device for constructing a capacity retention rate estimation model in one embodiment. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0051] The method for constructing a capacity retention rate estimation model provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown in FIG. 1 , the computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 1 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store construction data of a capacity retention rate estimation model. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for constructing a capacity retention rate estimation model is implemented.

[0052] Those skilled in the art will understand that Figure 1 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0053] In one embodiment, Figure 2As shown in FIG. 1 , a method for constructing a capacity retention rate estimation model is provided, and the method is applied to Figure 1 The following is an example of a server in the example, including:

[0054] S201, acquiring a plurality of feature data sets according to charge and discharge data of a plurality of sample batteries; each feature data set includes a plurality of sample features.

[0055] Among them, the sample battery can be a lithium-ion battery, and the charge and discharge data of the sample battery can be the long-term electrochemical performance data of the battery. For example, the charge and discharge data of the sample battery can be the data of a complete charge and discharge cycle of 50 to 500 weeks; multiple sample characteristics can include the maximum value, minimum value, difference, average value, variance of the charge and discharge voltage, and the change trend of each indicator with the increase of the number of charge and discharge cycles, the attenuation trend and variance of the capacity, the extreme points, change trend and n-th order distance of the differential capacity test curve, etc.

[0056] In an embodiment of the present application, the server may obtain the charge and discharge data of multiple sample batteries from a data acquisition device for sample batteries in advance, and store them in a local database; or, the server may obtain the charge and discharge data of multiple sample batteries from a data storage device. Furthermore, the charge and discharge data of multiple sample batteries are cleaned to remove irrelevant or redundant features in the charge and discharge data of multiple sample batteries, and missing values ​​and outliers are processed to obtain the charge and discharge data of each sample battery after cleaning, thereby performing feature extraction on the charge and discharge data of each sample battery after cleaning to obtain sample data of each sample battery, wherein the sample data of each battery includes multiple sample features, and then the sample data of all sample batteries are sampled and processed to generate multiple feature data sets. Optionally, the sample data of all batteries can be expressed as ,in is the sample data of the i-th sample battery, is the target variable of the i-th sample data, that is, the capacity retention rate. n sample data are randomly selected with replacement in multiple D times to obtain multiple feature data sets, that is, each feature data set includes n sample data.

[0057] For example, to remove irrelevant or redundant features in the charge and discharge data of multiple sample batteries, the battery number and features directly related to the later capacity retention rate can be deleted to avoid information leakage and model overfitting. To process missing values ​​and outliers, you can first check whether there are missing values ​​or outliers in the data. For missing values, you can use the method of deleting samples or interpolating values ​​to process them; for outliers, you can use statistical methods to detect and process them, such as using box plots or Z-score methods to detect and process them.

[0058] Optionally, feature standardization processing can be performed on multiple sample features in multiple feature data sets to eliminate the dimensional differences between different features, obtain multiple processed feature data sets, and use the multiple feature data sets to train the initial random forest model. Exemplarily, a standardization method can be used to convert each feature into data with a mean of 0 and a standard deviation of 1. For example, the standardization method can be shown in Formulas 1 to 3:

[0059] (Formula 1)

[0060] (Formula 2)

[0061] (Formula 3)

[0062] in, is the i-th feature in the data set and the j-th feature in the data set, for The corresponding standardized data is is the mean of the jth feature in all feature datasets, is the standard deviation of the jth feature, and n is the number of feature data sets, i.e., the number of sample batteries.

[0063] Optionally, feature extraction is performed on the charge and discharge data of multiple sample batteries to obtain multiple candidate sample features, and the correlation between each candidate sample feature and the capacity retention rate can be determined, thereby removing the candidate sample features whose correlation is not greater than the relevant threshold, retaining the candidate sample features whose correlation is greater than the relevant threshold, and obtaining multiple sample features to form multiple feature data sets. Exemplarily, the correlation between each candidate sample feature and the capacity retention rate can be determined based on the Pearson correlation coefficient between each candidate sample feature and the capacity retention rate. The data complexity of the model can be reduced by using data processing methods such as standardization and correlation analysis.

[0064] S202, training an initial random forest model according to multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery; the loss of the initial random forest model during the training process is determined according to the feature weight of each sample feature.

[0065] In the embodiment of the present application, at each node of the initial random forest model, a weighted probability sampling method is used to determine a feature subset. Specifically, at each node of the initial random forest model, according to the feature weight Construct a weight distribution, perform weighted random sampling on the sample features in the feature data set, select m sample features to form a feature subset S, further, determine the loss of node splitting based on the feature subset of the current node, and introduce the feature weight into the loss of node splitting to improve the model's sensitivity to important features. Optionally, the node splitting direction of each node can be determined based on the loss corresponding to the leaf node of each node. In an embodiment of the present application, node splitting is performed recursively until the stopping condition of node splitting is met to obtain a target prediction model. Optionally, the stopping condition can be that the random forest model reaches the maximum depth, or that the number of leaf node samples is less than the minimum number of samples, etc. Optionally, weighted probability sampling can be performed on multiple feature data sets to determine a feature subset; or, weighted probability sampling can be performed on any feature data set to determine a feature subset.

[0066] In the method for constructing the capacity retention rate estimation model, multiple feature data sets are obtained based on the charge and discharge data of multiple sample batteries; each feature data set includes multiple sample features; the initial random forest model is trained based on the multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery; the loss of the initial random forest model during the training process is determined based on the feature weights of each sample feature. The feature weights are introduced into the loss of node splitting to improve the sensitivity of the model to important features, thereby improving the efficiency of training the initial random forest model and improving the accuracy of the target prediction model; and the initial random forest model is trained based on multiple feature data sets, making full use of the nonlinear relationship between the sample features, thereby improving the prediction accuracy of the target prediction model for the later capacity retention rate of the battery.

[0067] In one embodiment, an implementation of the above S202 is provided, such as Figure 3 As shown, the above “training the initial random forest model according to multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery” includes:

[0068] S301, training an initial random forest model according to multiple feature data sets to obtain a prediction model corresponding to each feature data set.

[0069] In one embodiment, for each feature data set, at each node of the initial random forest model, a weighted probability sampling method is used to determine a feature subset. Furthermore, the loss of node splitting is determined based on the feature subset of the current node, and the direction of node splitting is determined based on the loss of each node until the stopping condition for node splitting is met, thereby obtaining a prediction model corresponding to each feature data set.

[0070] S302: construct a target prediction model for predicting the capacity retention rate of any target battery according to the prediction model corresponding to each feature data set and the model weight of each prediction model.

[0071] In the embodiment of the present application, if the number of prediction models is T, the target prediction model can be shown as Formula 4:

[0072] (Formula 4)

[0073] in, is the t-th prediction model, is the model weight of the t-th prediction model.

[0074] As an optional implementation, the weight of each feature data set may be preconfigured, so that the model weight of each prediction model is determined according to the weight of each feature data set, for example, the weight of the feature data set is determined as the model weight of the corresponding prediction model.

[0075] As another optional implementation, some of the feature data sets in the multiple feature data sets can be used as training sets to train multiple prediction models, and the remaining feature data sets in the multiple feature data sets can be used as test sets to perform performance tests on the multiple prediction models, and the model weights of the prediction models can be determined based on the performance test results of each prediction model. For example, the prediction accuracy obtained by the performance test of each prediction model is used as the model weight of each prediction model. Optionally, the performance of each prediction model in predicting the capacity retention rate of the 100th charge and discharge cycle in the training set is as follows: Figure 4 As shown in Figure 2, the root mean square error (RMSE) is 0.003403, the mean absolute percentage error (MAPE) is 0.25%, and the determination coefficient R 2 is 0.8779; the prediction set of the capacity retention rate prediction model of the 100th charge and discharge cycle of each prediction model is as follows Figure 5 As shown, the RMSE is 0.004519, the MAPE is 0.46%, and the R 2For example, at the 100th charge and discharge cycle of each sample, the actual value of the capacity retention rate of sample battery 0 is 0.941, and the predicted value of the capacity retention rate predicted by the target prediction model for sample battery 0 is 0.933; the actual value of the capacity retention rate of sample battery 1 is 0.917, and the predicted value of the capacity retention rate predicted by the target prediction model for sample battery 1 is 0.928; the actual value of the capacity retention rate of sample battery 2 is 0.938, and the predicted value of the capacity retention rate predicted by the target prediction model for sample battery 2 is 0.947; the actual value of the capacity retention rate of sample battery 3 is 0.953, and the predicted value of the capacity retention rate predicted by the target prediction model for sample battery 3 is 0.960. The predicted value of the capacity retention rate of sample battery 4 is 0.947; the actual value of the capacity retention rate of sample battery 4 is 0.935, and the predicted value of the capacity retention rate predicted by the target prediction model for sample battery 4 is 0.931; the actual value of the capacity retention rate of sample battery 5 is 0.949, and the predicted value of the capacity retention rate predicted by the target prediction model for sample battery 5 is 0.944; the actual value of the capacity retention rate of sample battery 6 is 0.926, and the predicted value of the capacity retention rate predicted by the target prediction model for sample battery 6 is 0.93; the actual value of the capacity retention rate of sample battery 7 is 0.943, and the predicted value of the capacity retention rate predicted by the target prediction model for sample battery 7 is 0.941.

[0076] In the above application embodiment, multiple prediction models are trained. Compared with training a single prediction model, the target prediction model obtained by multiple prediction models has higher accuracy, and the prediction accuracy of the target prediction model is further improved by combining the model weights of each prediction model.

[0077] In one embodiment, an implementation of the above S301 is provided, such as Figure 6 As shown, the above “training the initial random forest model according to multiple feature data sets to obtain the prediction model corresponding to each feature data set” includes:

[0078] S401, inputting sample features into an initial random forest model to obtain an initial prediction result; the initial random forest model is composed of a plurality of leaf nodes and a root node.

[0079] In an embodiment of the present application, for each set of feature data sets, the sample data in the feature data sets are input into the random forest model, so that the initial random forest model performs initial node splitting from the root node according to the sample features to form leaf nodes, and predicts the initial prediction results.

[0080] S402, determining the node loss of each leaf node in the initial random forest model according to the feature weight of each sample feature and the initial prediction result.

[0081] In an embodiment of the present application, the weighted minimum mean square error of each leaf node and the weighted minimum mean square error of the child nodes of each leaf node that undergoes node splitting can be determined based on the feature weights of each sample feature, thereby determining the node loss of the leaf node based on the weighted minimum mean square error of the leaf node and the weighted minimum mean square error of the child nodes.

[0082] Optional, such as Figure 7 As shown, the process of determining the feature weight of each sample feature may include:

[0083] S404, determining the correlation between each sample characteristic and the capacity retention rate.

[0084] S405: Determine a sample weight of each sample feature according to the correlation between each sample feature and the capacity retention rate.

[0085] In the embodiment of the present application, the correlation between each sample feature and the capacity retention rate can be determined based on the Pearson correlation coefficient between each sample feature and the capacity retention rate. Further, the sum of all correlations can be determined first to obtain the sum of correlations, and then the ratio between the correlation between each sample feature and the capacity retention rate and the sum of correlations can be determined as the sample weight corresponding to each sample feature.

[0086] S403, training the initial random forest model according to the loss of each node to obtain a prediction model corresponding to each feature data set.

[0087] In an embodiment of the present application, for the splitting of the root node and each leaf node, the sample features and splitting points corresponding to the maximum node loss are selected for node splitting, the leaf nodes after the split are determined, and the above-mentioned step of "inputting the sample features into the initial random forest model to obtain the initial prediction results" is returned to execute until the stopping condition of the node splitting is met, and the prediction model corresponding to each feature data set is obtained.

[0088] In the above application embodiment, feature weights are introduced into the loss of node splitting to improve the sensitivity of the model to important features, thereby improving the efficiency of training the initial random forest model and improving the accuracy of the target prediction model.

[0089] In one embodiment, an implementation of the above S403 is provided, such as Figure 8 As shown in the figure, the above “determining the node loss of each leaf node in the initial random forest model according to the feature weight of each sample feature and the initial prediction result” includes:

[0090] S501, determining a first weighted mean square error of each leaf node in the initial random forest model and a second weighted mean square error of a child node associated with each leaf node according to a feature weight of each sample feature.

[0091] In the embodiment of the present application, the first weighted mean square error of each leaf node is determined according to the feature weight of each sample feature, and the second weighted mean square error of the child node associated with each leaf node is determined according to the feature weight of the sample feature. The first weighted mean square error and the second weighted mean square error are determined in the same manner. Exemplarily, the first weighted mean square error of each leaf node can be shown as Formula 5:

[0092] (Formula 5)

[0093] in, is the number of samples of node N, is the predicted value of sample data i, is the sample weight of the jth sample feature.

[0094] S502, determining the node loss of each leaf node in the initial random forest model according to the first weighted mean square error of each leaf node and the second weighted mean square error of each child node.

[0095] In the embodiment of the present application, the node loss of each leaf node can be expressed as Formula 6:

[0096] (Formula 6)

[0097] in, is the first weighted mean square error of node N, is the left child of node N, is the right child of node N, For Node The number of samples, For Node The number of samples, For child nodes The second weighted mean square error, For child nodes The second weighted mean square error.

[0098] In the above application embodiment, each leaf node and the weighted mean square error of the child nodes associated with each leaf node are determined according to the feature weights of the sample features, so that the node splitting of the random forest model is more matched with the target battery being tested, thereby improving the accuracy of the target prediction model in predicting the capacity retention rate of the battery.

[0099] In one embodiment, Fig. 9 As shown, the method for constructing the above capacity retention rate estimation model also includes:

[0100] S203, acquiring charging and discharging data of the target battery, and determining a feature data set corresponding to the target battery according to the charging and discharging data.

[0101] Among them, the target battery can be a lithium-ion battery, and the charge and discharge data of the target battery can be the early electrochemical performance data of the battery. For example, the charge and discharge data of the target battery can be the data of the complete charge and discharge cycle of the first 50 weeks; the characteristic data set can include the maximum value, minimum value, difference, average value, variance of the charge and discharge voltage, and the transformation trend of each indicator with the increase of the number of charge and discharge cycles, the attenuation trend and variance of the capacity, the extreme points, change trend and n-th order distance of the differential capacity test curve, etc.

[0102] In an embodiment of the present application, data cleaning is performed on the charge and discharge data of the target battery to remove irrelevant or redundant features in the charge and discharge data of the target battery, and missing values ​​and outliers are processed to obtain cleaned charge and discharge data, thereby performing feature extraction on the cleaned charge and discharge data to obtain a feature data set.

[0103] Optionally, feature extraction is performed on the charge and discharge data of the target battery to obtain multiple candidate features. The candidate features of the target battery are screened according to the correlation between each sample feature of the sample battery and the capacity retention rate to obtain the screened candidate features and form a feature data set.

[0104] S204, inputting the characteristic data set corresponding to the target battery into the target prediction model, and outputting the capacity retention rate of the target battery.

[0105] In an embodiment of the present application, the target prediction model includes multiple trained random forest models. The feature data set is input into the multiple trained random forest models in the target prediction model for prediction, and the prediction results of each random forest model are obtained. The prediction results are then weighted averaged to obtain the capacity retention rate of the target battery.

[0106] In the embodiments of the present application, the early electrochemical performance data of the battery can be used to predict its later capacity retention rate, without the need for long-term cycle testing, thus shortening the time for battery performance testing.

[0107] In one embodiment, a method for constructing a complete capacity retention rate estimation model is provided, such as Fig.10 As shown, the above method includes:

[0108] S1, obtaining a plurality of feature data sets according to charge and discharge data of a plurality of sample batteries; each feature data set includes a plurality of sample features.

[0109] S2, determine the correlation between each sample characteristic and capacity retention rate.

[0110] S3, determining the sample weight of each sample feature according to the correlation between each sample feature and the capacity retention rate.

[0111] S4, input the sample features into the initial random forest model to obtain the initial prediction results; the initial random forest model consists of multiple leaf nodes and a root node.

[0112] S5, determining a first weighted mean square error of each leaf node in the initial random forest model and a second weighted mean square error of a child node associated with each leaf node according to a feature weight of each sample feature.

[0113] S6, determining the node loss of each leaf node in the initial random forest model according to the first weighted mean square error of each leaf node and the second weighted mean square error of each child node.

[0114] S7, train the initial random forest model according to the loss of each node to obtain the prediction model corresponding to each feature data set.

[0115] S8, constructing a target prediction model for predicting the capacity retention rate of any target battery according to the prediction model corresponding to each feature data set and the model weight of each prediction model.

[0116] S9, acquiring charging and discharging data of the target battery, and determining a feature data set corresponding to the target battery according to the charging and discharging data.

[0117] S10, inputting the characteristic data set corresponding to the target battery into the target prediction model, and outputting the capacity retention rate of the target battery.

[0118] In the method for constructing the capacity retention rate estimation model, multiple feature data sets are obtained based on the charge and discharge data of multiple sample batteries; each feature data set includes multiple sample features; the initial random forest model is trained based on the multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery; the loss of the initial random forest model during the training process is determined based on the feature weights of each sample feature. The feature weights are introduced into the loss of node splitting to improve the sensitivity of the model to important features, thereby improving the efficiency of training the initial random forest model and improving the accuracy of the target prediction model; and the initial random forest model is trained based on multiple feature data sets, making full use of the nonlinear relationship between the sample features, thereby improving the prediction accuracy of the target prediction model for the later capacity retention rate of the battery.

[0119] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0120] Based on the same inventive concept, the embodiment of the present application also provides a device for constructing a capacity retention rate estimation model for implementing the method for constructing a capacity retention rate estimation model involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiments of the device for constructing one or more capacity retention rate estimation models provided below can be referred to the limitations of the method for constructing a capacity retention rate estimation model above, and will not be repeated here.

[0121] In one embodiment, Fig.11 As shown, a device for constructing a capacity retention rate estimation model is provided, comprising: a first acquisition module 10 and a construction module 11, wherein:

[0122] The first acquisition module 10 is used to acquire a plurality of feature data sets according to the charge and discharge data of a plurality of sample batteries; each feature data set includes a plurality of sample features.

[0123] Construction module 11 is used to train the initial random forest model according to multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery; the loss of the initial random forest model during the training process is determined according to the feature weight of each sample feature.

[0124] In one embodiment, the construction module 11 includes: a training unit and a construction unit, wherein:

[0125] The training unit is used to train the initial random forest model according to multiple feature data sets to obtain a prediction model corresponding to each feature data set.

[0126] The construction unit is used to construct a target prediction model for predicting the capacity retention rate of any target battery according to the prediction model corresponding to each feature data set and the model weight of each prediction model.

[0127] In one embodiment, the above-mentioned training unit is specifically used to input sample features into an initial random forest model to obtain an initial prediction result; the initial random forest model is composed of multiple leaf nodes and a root node; according to the feature weights of each sample feature and the initial prediction result, the node loss of each leaf node in the initial random forest model is determined; the initial random forest model is trained according to each node loss to obtain a prediction model corresponding to each feature data set.

[0128] In one embodiment, the above-mentioned training unit is specifically used to determine the first weighted mean square error of each leaf node in the initial random forest model and the second weighted mean square error of the child nodes associated with each leaf node according to the feature weights of each sample feature; and determine the node loss of each leaf node in the initial random forest model according to the first weighted mean square error of each leaf node and the second weighted mean square error of each child node.

[0129] In one embodiment, the training unit is specifically used to determine the correlation between each sample feature and the capacity retention rate; and determine the sample weight of each sample feature according to the correlation between each sample feature and the capacity retention rate.

[0130] In one embodiment, the device for constructing the capacity retention rate estimation model further includes: a second acquisition module and an output module, wherein:

[0131] The second acquisition module is used to acquire the charge and discharge data of the target battery, and determine the characteristic data set corresponding to the target battery according to the charge and discharge data.

[0132] The output module is used to input the characteristic data set corresponding to the target battery into the target prediction model, and output the capacity retention rate of the target battery.

[0133] Each module in the above-mentioned device for constructing the capacity retention rate estimation model can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0134] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0135] Acquire multiple feature data sets according to the charge and discharge data of multiple sample batteries; each feature data set includes multiple sample features;

[0136] The initial random forest model is trained according to multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery; the loss of the initial random forest model during the training process is determined according to the feature weight of each sample feature.

[0137] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0138] The initial random forest model is trained according to multiple feature data sets to obtain the prediction model corresponding to each feature data set;

[0139] According to the prediction models corresponding to each feature data set and the model weights of each prediction model, a target prediction model for predicting the capacity retention rate of any target battery is constructed.

[0140] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0141] Input the sample features into the initial random forest model to obtain the initial prediction results; the initial random forest model consists of multiple leaf nodes and a root node;

[0142] Determine the node loss of each leaf node in the initial random forest model based on the feature weights of each sample feature and the initial prediction results;

[0143] The initial random forest model is trained according to the loss of each node to obtain the prediction model corresponding to each feature data set.

[0144] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0145] Determine, according to the feature weights of the features of each sample, a first weighted mean square error of each leaf node in the initial random forest model and a second weighted mean square error of the child nodes associated with each leaf node;

[0146] The node loss of each leaf node in the initial random forest model is determined according to the first weighted mean square error of each leaf node and the second weighted mean square error of each child node.

[0147] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0148] Determine the correlation between each sample characteristic and capacity retention;

[0149] According to the correlation between each sample feature and the capacity retention rate, the sample weight of each sample feature is determined.

[0150] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0151] Acquire charge and discharge data of the target battery, and determine a feature data set corresponding to the target battery according to the charge and discharge data;

[0152] The characteristic data set corresponding to the target battery is input into the target prediction model, and the capacity retention rate of the target battery is obtained as output.

[0153] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0154] Acquire multiple feature data sets according to the charge and discharge data of multiple sample batteries; each feature data set includes multiple sample features;

[0155] The initial random forest model is trained according to multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery; the loss of the initial random forest model during the training process is determined according to the feature weight of each sample feature.

[0156] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0157] The initial random forest model is trained according to multiple feature data sets to obtain the prediction model corresponding to each feature data set;

[0158] According to the prediction models corresponding to each feature data set and the model weights of each prediction model, a target prediction model for predicting the capacity retention rate of any target battery is constructed.

[0159] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0160] Input the sample features into the initial random forest model to obtain the initial prediction results; the initial random forest model consists of multiple leaf nodes and a root node;

[0161] Determine the node loss of each leaf node in the initial random forest model based on the feature weights of each sample feature and the initial prediction results;

[0162] The initial random forest model is trained according to the loss of each node to obtain the prediction model corresponding to each feature data set.

[0163] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0164] Determine, according to the feature weights of the features of each sample, a first weighted mean square error of each leaf node in the initial random forest model and a second weighted mean square error of the child nodes associated with each leaf node;

[0165] The node loss of each leaf node in the initial random forest model is determined according to the first weighted mean square error of each leaf node and the second weighted mean square error of each child node.

[0166] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0167] Determine the correlation between each sample characteristic and capacity retention;

[0168] According to the correlation between each sample feature and the capacity retention rate, the sample weight of each sample feature is determined.

[0169] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0170] Acquire charge and discharge data of the target battery, and determine a feature data set corresponding to the target battery according to the charge and discharge data;

[0171] The characteristic data set corresponding to the target battery is input into the target prediction model, and the capacity retention rate of the target battery is obtained as output.

[0172] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0173] Acquire multiple feature data sets according to the charge and discharge data of multiple sample batteries; each feature data set includes multiple sample features;

[0174] The initial random forest model is trained according to multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery; the loss of the initial random forest model during the training process is determined according to the feature weight of each sample feature.

[0175] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0176] The initial random forest model is trained according to multiple feature data sets to obtain the prediction model corresponding to each feature data set;

[0177] According to the prediction models corresponding to each feature data set and the model weights of each prediction model, a target prediction model for predicting the capacity retention rate of any target battery is constructed.

[0178] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0179] Input the sample features into the initial random forest model to obtain the initial prediction results; the initial random forest model consists of multiple leaf nodes and a root node;

[0180] Determine the node loss of each leaf node in the initial random forest model based on the feature weights of each sample feature and the initial prediction results;

[0181] The initial random forest model is trained according to the loss of each node to obtain the prediction model corresponding to each feature data set.

[0182] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0183] Determine, according to the feature weights of the features of each sample, a first weighted mean square error of each leaf node in the initial random forest model and a second weighted mean square error of the child nodes associated with each leaf node;

[0184] The node loss of each leaf node in the initial random forest model is determined according to the first weighted mean square error of each leaf node and the second weighted mean square error of each child node.

[0185] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0186] Determine the correlation between each sample characteristic and capacity retention;

[0187] According to the correlation between each sample feature and the capacity retention rate, the sample weight of each sample feature is determined.

[0188] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0189] Acquire charge and discharge data of the target battery, and determine a feature data set corresponding to the target battery according to the charge and discharge data;

[0190] The characteristic data set corresponding to the target battery is input into the target prediction model, and the capacity retention rate of the target battery is obtained as output.

[0191] It should be noted that the data involved in this application (including but not limited to data used for analysis, stored data, displayed data, etc.) are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0192] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0193] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0194] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the present application. It should be noted that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for constructing a capacity retention rate estimation model, characterized in that: The method comprises: Acquire multiple feature data sets according to the charge and discharge data of multiple sample batteries; each of the feature data sets includes multiple sample features; The initial random forest model is trained according to the multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery; the loss of the initial random forest model during the training process is determined according to the feature weight of each of the sample features.

2. The method according to claim 1, characterized in that The initial random forest model is trained according to the multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery, including: Training the initial random forest model according to the multiple feature data sets to obtain a prediction model corresponding to each feature data set; According to the prediction models corresponding to the characteristic data sets and the model weights of the prediction models, a target prediction model for predicting the capacity retention rate of any target battery is constructed.

3. The method according to claim 2, characterized in that The initial random forest model is trained according to the multiple feature data sets to obtain a prediction model corresponding to each feature data set, including: Inputting the sample features into the initial random forest model to obtain an initial prediction result; the initial random forest model is composed of a plurality of leaf nodes and a root node; Determining the node loss of each leaf node in the initial random forest model according to the feature weight of each sample feature and the initial prediction result; The initial random forest model is trained according to the node losses to obtain the prediction model corresponding to each feature data set.

4. The method according to claim 3, characterized in that Determining the node loss of each leaf node in the initial random forest model according to the feature weight of each sample feature and the initial prediction result includes: Determine, according to the feature weights of the sample features, a first weighted mean square error of each leaf node in the initial random forest model and a second weighted mean square error of a child node associated with each leaf node; The node loss of each leaf node in the initial random forest model is determined according to the first weighted mean square error of each leaf node and the second weighted mean square error of each child node.

5. The method according to claim 3, characterized in that: The method further comprises: Determining the correlation between each of the sample characteristics and the capacity retention rate; According to the correlation between each of the sample characteristics and the capacity retention rate, the sample weight of each of the sample characteristics is determined.

6. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: Acquiring charging and discharging data of the target battery, and determining a feature data set corresponding to the target battery according to the charging and discharging data; The characteristic data set corresponding to the target battery is input into the target prediction model, and the capacity retention rate of the target battery is obtained as an output.

7. A device for constructing a capacity retention rate estimation model, characterized in that: The device comprises: A first acquisition module, configured to acquire a plurality of feature data sets according to charge and discharge data of a plurality of sample batteries; each of the feature data sets includes a plurality of sample features; A construction module is used to train an initial random forest model according to the multiple feature data sets to construct a target prediction model for predicting the capacity retention rate of any target battery; the loss of the initial random forest model during the training process is determined according to the feature weight of each of the sample features.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images