A lithium battery life prediction method and system based on graph neural network

Abstract

The present invention discloses a lithium battery life prediction method and system based on a graph neural network. The method includes: obtaining charge and discharge cycle data of several historical periods of the lithium battery, extracting battery capacity data based on the charge and discharge cycle data; segmenting the battery capacity data to obtain multiple sequence samples corresponding to the time series, and dividing the sequence samples into training sets and test sets according to preset time nodes to train a preset neural network model; adaptively adding white noise to the battery capacity data to amplify the degree of non-correlation between each mode to decompose the mode of the battery capacity data; constructing a graph neural network, and combining the adjacency matrix, graph convolution and attention mechanism to obtain a lithium battery life prediction model, and predicting the remaining life of the lithium battery through the lithium battery life prediction model. The present invention solves the problem of insufficient lithium battery life prediction accuracy and limited improvement in prediction accuracy in the prior art.

Description

Technical Field

[0001] The present invention relates to the technical field of battery life prediction, and in particular to a lithium battery life prediction method and system based on graph neural network. Background Art

[0002] While existing battery life prediction technologies have improved prediction accuracy to a certain extent, they still have some shortcomings. First, many methods fail to adequately model the complex dependencies between modal components and fail to fully exploit the synergies between them. Second, feature fusion and information integration methods are relatively simple and cannot effectively address feature dispersion and noise. Finally, some methods rely on manually set parameters or rules, lacking adaptability and difficulty adapting to the variability and complexity of diverse data.

[0003] Existing methods for predicting the remaining life of lithium batteries typically use modal decomposition techniques, such as CEEMDAN, to decompose battery capacity data into multiple modal components to extract different feature information. Although existing methods have improved prediction accuracy to a certain extent, several key issues still exist. First, many methods do not adequately model the complex dependencies between modal components and are unable to fully explore the synergy between components, resulting in limited improvement in prediction accuracy. Second, the feature fusion and information integration methods are relatively simple and fail to effectively address the problem of feature dispersion, affecting the comprehensive understanding of the battery degradation process. Finally, some methods rely on manually set parameters or rules, lack adaptability, and cannot flexibly adapt to the changes and complexity of different data sets.

[0004] While existing battery life prediction technologies have improved prediction accuracy to a certain extent, they still have some shortcomings. First, many methods fail to adequately model the complex dependencies between modal components and fail to fully exploit the synergies between them. Second, feature fusion and information integration methods are relatively simple and cannot effectively address feature dispersion and noise. Finally, some methods rely on manually set parameters or rules, lacking adaptability and difficulty adapting to the variability and complexity of diverse data. Summary of the Invention

[0005] In view of the shortcomings of the existing technology, the purpose of the present invention is to provide a lithium battery life prediction method and system based on graph neural network, aiming to solve the problems in the existing technology of insufficient lithium battery life prediction accuracy and limited improvement of prediction accuracy.

[0006] A first aspect of the present invention is to provide a lithium battery life prediction method based on a graph neural network, the method comprising:

[0007] Acquire charge and discharge cycle data of several historical periods of the lithium battery, and extract battery capacity data based on the charge and discharge cycle data;

[0008] Segmenting the battery capacity data to obtain a plurality of sequence samples corresponding to a time series, and dividing the sequence samples into a training set and a test set according to preset time nodes to train a preset neural network model;

[0009] Adaptively adding white noise to the battery capacity data to amplify the degree of non-correlation between modes, so as to decompose the modes of the battery capacity data;

[0010] A graph neural network is constructed and combined with the adjacency matrix, graph convolution and attention mechanism to obtain a lithium battery life prediction model, which is used to predict the remaining life of the lithium battery.

[0011] According to one aspect of the above technical solution, the steps of segmenting the battery capacity data to obtain a plurality of sequence samples corresponding to time series, and dividing the sequence samples into a training set and a test set according to preset time nodes to train a preset neural network model include:

[0012] Continuous battery capacity data is segmented into sliding windows of fixed length. Each fixed-length data segment is used as a sequence sample to capture the dynamic changes in the time series. The target value of each sequence sample is the data of the next time step in the time series, which is used to predict future capacity values.

[0013] Dividing the plurality of sequence samples according to a preset ratio in chronological order to obtain a training set for training a neural network model and a test set for testing the neural network model;

[0014] The neural network model is trained using the training set and the test set.

[0015] According to one aspect of the above technical solution, the step of adaptively adding white noise to the battery capacity data to amplify the degree of non-correlation between the modes to decompose the modes of the battery capacity data includes:

[0016] Adding white noise to the original battery capacity data using an integrated empirical mode decomposition method based on adaptive white noise;

[0017] The degree of non-correlation between the modes in the battery capacity data is amplified by adding white noise, so as to perform modal decomposition on the capacity sequence in the battery capacity data.

[0018] According to one aspect of the above technical solution, a graph neural network is constructed, and an adjacency matrix, graph convolution, and attention mechanism are combined to characterize the complex relationship between modal components, thereby obtaining a lithium battery life prediction model to predict the remaining life of the lithium battery through the lithium battery life prediction model, including the following steps:

[0019] Obtaining the time series;

[0020] Converting multiple modal components obtained based on capacity sequence decomposition into multiple graph nodes in a modal component graph, and using an adjacency matrix to represent the relationship between the multiple graph nodes in the modal component graph;

[0021] Obtaining the structure of the modal component graph at each moment, employing a graph attention layer to capture the relationship between modal components using an attention mechanism, calculating the attention weights between a modal component and its associated modal components, updating the current modal component state based on the attention weights to extract node spatial features between multiple modal components, and aggregating the node spatial features using average pooling to obtain global spatial features;

[0022] According to the global spatial features, a gated recurrent unit is used to construct a feature matrix of the time series, and the input time series is modeled through a gating mechanism;

[0023] The features processed by the graph attention layer and the gated recurrent unit are input into the fully connected layer to predict and output the remaining life of the lithium battery.

[0024] According to one aspect of the above technical solution, in the step of constructing a feature matrix of a time series using a gated recurrent unit according to the global spatial features, and modeling the input time series through a gating mechanism, the step of modeling the input time series through a gating mechanism using a gated recurrent unit includes:

[0025] a. Input: Controls the global spatial characteristics of the current input Current status The influence of is expressed as:

[0026] ;

[0027] Where, represents the reset gate vector, represents the activation function, and Represent the weight and bias of the gated recurrent unit respectively;

[0028] b. Update gate: control the previous step state and the global spatial features of the current input The degree of fusion is expressed as:

[0029] ;

[0030] Where, To update the gate vector, To update the gate weights, To update the gate bias;

[0031] c. Candidate state calculation: Generate candidate states at the current moment , the expression is:

[0032] ;

[0033] Where, is the candidate state weight, is the candidate state bias;

[0034] d. State update: Combined update gate vector With candidate status , calculate the final state at the current moment , the expression is:

[0035] ;

[0036] e. Every moment , the gated recurrent unit will output a hidden state vector, namely , represents the current moment feature, the last hidden state of the gated recurrent unit It is used as the final feature representation for lithium battery life prediction.

[0037] According to one aspect of the above technical solution, the features processed by the graph attention layer and the gated recurrent unit are input into the fully connected layer to predict and output the remaining life of the lithium battery. The expression is:

[0038] ;

[0039] Where, is the predicted value of the remaining life of the lithium battery, is the weight of the fully connected layer, is the bias of the fully connected layer.

[0040] According to one aspect of the above technical solution, the method further includes:

[0041] The mean absolute percentage error is used as an evaluation index to evaluate the remaining life prediction value output by the lithium battery life prediction model. The expression is:

[0042] ;

[0043] Where, is the actual value of the remaining life of the lithium battery, is the predicted value of the remaining life of the lithium battery, is the sample size.

[0044] A second aspect of the present invention is to provide a lithium battery life prediction system based on a graph neural network, which is applied to the method described in the above technical solution. The system includes:

[0045] A data extraction module is used to obtain charge and discharge cycle data of several historical periods of the lithium battery and extract battery capacity data based on the charge and discharge cycle data;

[0046] a sample segmentation module, configured to segment the battery capacity data to obtain a plurality of sequence samples corresponding to time series, and to divide the sequence samples into a training set and a test set according to preset time nodes, so as to train a preset neural network model;

[0047] a modal decomposition module, configured to adaptively add white noise to the battery capacity data to amplify the degree of non-correlation between modes, thereby decomposing the modes of the battery capacity data;

[0048] The life prediction module is used to construct a graph neural network and combine the adjacency matrix, graph convolution and attention mechanism to obtain a lithium battery life prediction model, and predict the remaining life of the lithium battery through the lithium battery life prediction model.

[0049] A third aspect of the present invention is to provide a readable storage medium having a computer program stored thereon, which implements the method described in the above technical solution when executed by a processor.

[0050] The fourth aspect of the present invention is to provide an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in the above technical solution when executing the computer program.

[0051] Compared with the prior art, the lithium battery life prediction method and system based on graph neural network shown in the present invention has the following beneficial effects:

[0052] The method presented in this paper eliminates the reliance on manually set parameters and rules and instead employs adaptive learning and feature integration mechanisms, enabling the lithium battery life prediction model to flexibly adapt to the variability and complexity of diverse datasets, improving its stability and prediction accuracy in a variety of complex situations. This fusion of spatiotemporal features not only enhances the synergy between modal components but also significantly improves the accuracy and effectiveness of prediction results, providing an efficient and highly adaptable solution for accurately predicting the remaining life of lithium batteries. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments with reference to the accompanying drawings, in which:

[0054] Figure 1 A schematic diagram of a flow chart of a lithium battery life prediction method based on a graph neural network provided in an embodiment of the present invention;

[0055] Figure 2 This is a structural block diagram of a lithium battery life prediction system based on graph neural network provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0056] To make the objectives, features, and advantages of the present invention more readily apparent, the following detailed description of specific embodiments of the present invention is provided in conjunction with the accompanying drawings. The accompanying drawings illustrate several embodiments of the present invention. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and comprehensive understanding of the present invention.

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one skilled in the art to which this invention pertains. The terms used in this specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0058] Example 1

[0059] See also Figure 1 The first embodiment of the present invention provides a lithium battery life prediction method based on a graph neural network. The method shown in this embodiment includes steps S10 to S40:

[0060] Step S10: acquiring charge and discharge cycle data of several historical periods of the lithium battery, and extracting battery capacity data based on the charge and discharge cycle data.

[0061] In this embodiment, the raw discharge cycle data of several historical cycles of the lithium battery are obtained, specifically based on the historical cycle charge and discharge data of the lithium battery at 24°C provided by the lithium battery NASA dataset, and the capacity index is extracted based on the discharge process as the battery health life feature.

[0062] In step S20, the battery capacity data is segmented to obtain a plurality of sequence samples corresponding to time series, and the sequence samples are divided into a training set and a test set according to preset time nodes to train a preset neural network model.

[0063] In this embodiment, the battery capacity data is segmented to obtain a plurality of sequence samples corresponding to time series, and the sequence samples are divided into a training set and a test set according to preset time nodes to train a preset neural network model, including the following steps:

[0064] Continuous battery capacity data is segmented into sliding windows of fixed length. Each fixed-length data segment is used as a sequence sample to capture the dynamic changes in the time series. The target value of each sequence sample is the data of the next time step in the time series, which is used to predict future capacity values.

[0065] Dividing the plurality of sequence samples according to a preset ratio in chronological order to obtain a training set for training a neural network model and a test set for testing the neural network model;

[0066] The neural network model is trained using the training set and the test set.

[0067] Specifically, step S20 mainly pre-processes the data, including:

[0068] (1) Sequence construction: The continuous battery capacity data is divided into sliding windows of fixed length T. Each segment of length T is used as a sequence sample to capture the dynamic change characteristics of the time series. The target value of each sample is the data of the next time step in the sequence, which is used to predict the future capacity value.

[0069] (2) Training set and test set division: The data is divided into two parts according to the time sequence. The first 80% of the data is used for training data, and the last 20% is used for testing the performance of the model, ensuring the continuity between the training set and the test set. The neural network model is trained using the training set and test set obtained by the division.

[0070] Step S30 : Adaptively adding white noise to the battery capacity data to amplify the degree of non-correlation between modes, so as to decompose the modes of the battery capacity data.

[0071] In this embodiment, the step of adaptively adding white noise to the battery capacity data to amplify the degree of non-correlation between modes to decompose the modes of the battery capacity data includes:

[0072] Adding white noise to the original battery capacity data using an integrated empirical mode decomposition method based on adaptive white noise;

[0073] The degree of non-correlation between the modes in the battery capacity data is amplified by adding white noise, so as to perform modal decomposition on the capacity sequence in the battery capacity data.

[0074] Specifically, the method of integrated empirical mode decomposition (EMD) based on adaptive white noise decomposes battery capacity data, which is often used to process non-stationary and nonlinear data. By adding white noise to the original data, it can amplify the degree of independence between the modes and facilitate the modal decomposition of the original data.

[0075] Step S40: construct a graph neural network, and combine the adjacency matrix, graph convolution and attention mechanism to obtain a lithium battery life prediction model, and predict the remaining life of the lithium battery through the lithium battery life prediction model.

[0076] In this embodiment, a graph neural network is constructed, and an adjacency matrix, graph convolution, and attention mechanism are combined to depict the complex relationship between modal components, thereby obtaining a lithium battery life prediction model to predict the remaining life of the lithium battery through the lithium battery life prediction model. The steps include:

[0077] Obtaining the time series;

[0078] Converting multiple modal components obtained based on capacity sequence decomposition into multiple graph nodes in a modal component graph, and using an adjacency matrix to represent the relationship between the multiple graph nodes in the modal component graph;

[0079] Obtaining the structure of the modal component graph at each moment, employing a graph attention layer to capture the relationship between modal components using an attention mechanism, calculating the attention weights between a modal component and its associated modal components, updating the current modal component state based on the attention weights to extract node spatial features between multiple modal components, and aggregating the node spatial features using average pooling to obtain global spatial features;

[0080] According to the global spatial features, a gated recurrent unit is used to construct a feature matrix of the time series, and the input time series is modeled through a gating mechanism;

[0081] The features processed by the graph attention layer and the gated recurrent unit are input into the fully connected layer to predict and output the remaining life of the lithium battery.

[0082] Specifically, the core innovation of this embodiment lies in introducing a dynamic graph neural network to construct dependencies between modal components, effectively integrating scattered information into feature representations with higher expressive power, and significantly enhancing the ability of the lithium battery life prediction model to capture battery degradation laws.

[0083] This lithium battery life prediction model achieves prediction tasks by combining time and space. A graph attention layer processes the spatial relationship of each time step, capturing the dependencies between modal components. A gated recurrent unit models time series features, ultimately achieving the integration and prediction of spatiotemporal features. The specific steps are as follows:

[0084] After modal decomposition of the capacity sequence, M modal component sequences are obtained As features of graph nodes.

[0085] 1. Input time series , , where T is the size of the time window.

[0086] 2. Build graph structure

[0087] (1) Convert the modal components into nodes in the graph

[0088] Every moment Construct the modal component diagram, and the node feature matrix is:

[0089] ;

[0090] in, .

[0091] (2) Constructing the adjacency matrix

[0092] The adjacency matrix is ​​used to represent the relationship between nodes in a graph. In battery life prediction, the Pearson correlation coefficient between modal components is calculated as an element of the adjacency matrix, which represents the similarity and dependency between different modal components.

[0093] M modal components , each component has T The values ​​at each time point form a matrix A :

[0094] ;

[0095] in, Represents the modal component i At the moment t The correlation coefficient is then obtained using the Pearson formula.

[0096] 3. Graph Attention Layer (GAT)

[0097] Enter each moment t The modal component graph structure of GAT uses the attention mechanism to capture the relationship between modal components. By calculating the attention weight between a modal component and its associated modal components, the current modal component state is updated according to the weight, thereby extracting the spatial features between multiple modal components.

[0098] The formula for updating node features is:

[0099] ;

[0100] in, yes l The node feature matrix of the layer, is a normalized adjacency matrix that represents the connection relationship between nodes in the graph. is the weight matrix of the current layer, is an activation function, such as Activation function. The node features are then aggregated using average pooling to obtain a global feature representation, providing a unified input for the GRU.

[0101] ;

[0102] in, N is the number of nodes, It is i The characteristics of a node.

[0103] 4. GRU module (Gated Recurrent Unit)

[0104] The GRU has a simpler structure and fewer parameters, enabling it to converge quickly when processing sequences. It can quickly learn short-term patterns and trends in time series and is particularly sensitive to frequently changing short-term fluctuations, making it well-suited to capturing battery regeneration in a short period of time.

[0105] (1) Every moment t The graph passes through the GAT module and outputs a global feature vector , the vector can be expressed as:

[0106] ;

[0107] in, N represents the number of modal components, It is t The moment i The value of the node feature after being updated and pooled by GAT.

[0108] (2) Time series construction

[0109] Every moment tThe global eigenvector of Considered as an input feature of the time series, for the input of T time steps, a feature matrix can be constructed:

[0110] ;

[0111] in, , T is the number of time steps, d is the dimension of the feature vector.

[0112] (3) GRU models the input time series through a gating mechanism. The following are the specific steps of GRU:

[0113] a. Input: Controls the global spatial characteristics of the current input Current status The influence of is expressed as:

[0114] ;

[0115] Where, represents the reset gate vector, represents the activation function, and Represent the weight and bias of the gated recurrent unit respectively;

[0116] b. Update gate: control the previous step state and the global spatial features of the current input The degree of fusion is expressed as:

[0117] ;

[0118] Where, To update the gate vector, To update the gate weights, To update the gate bias;

[0119] c. Candidate state calculation: Generate candidate states at the current moment , the expression is:

[0120] ;

[0121] Where, is the candidate state weight, is the candidate state bias;

[0122] d. State update: Combined update gate vector With candidate status , calculate the final state at the current moment , the expression is:

[0123] ;

[0124] e. Every moment , the gated recurrent unit will output a hidden state vector, namely , represents the current moment feature, the last hidden state of the gated recurrent unit It is used as the final feature representation for lithium battery life prediction.

[0125] In this embodiment, the features processed by the graph attention layer and the gated recurrent unit are input into the fully connected layer to predict and output the remaining life of the lithium battery. The expression is:

[0126] ;

[0127] Where, is the predicted value of the remaining life of the lithium battery, is the weight of the fully connected layer, is the bias of the fully connected layer.

[0128] In addition, in this embodiment, the method further includes:

[0129] The mean absolute percentage error is used as an evaluation index to evaluate the remaining life prediction value output by the lithium battery life prediction model. The expression is:

[0130] ;

[0131] Where, is the actual value of the remaining life of the lithium battery, is the predicted value of the remaining life of the lithium battery, is the sample size.

[0132] The present invention successfully captures the spatiotemporal features in the prediction of the remaining life of lithium batteries by combining the graph attention layer and the GRU unit, thereby significantly improving the accuracy and robustness of the prediction. The graph attention layer can accurately capture the dependencies between modal components by adaptively processing the spatial relationship of each time step, solving the problems of scattered modes and inaccurate capture that occur in traditional methods when dealing with capacity regeneration phenomena. In addition, the graph attention layer also enhances the influence of important features by adjusting the weights of features, while reducing the interference of false components on the prediction results, effectively reflecting the changes in battery status. Finally, the gated recurrent unit effectively captures the changing trends of battery life in different time periods by learning the dynamic laws of time series, further enhancing the understanding of the capacity regeneration phenomenon.

[0133] Compared to existing technologies, the method shown in this embodiment abandons reliance on manually set parameters and rules and instead employs adaptive learning and feature integration mechanisms, enabling the lithium battery life prediction model to flexibly respond to the variability and complexity of diverse datasets, improving its stability and prediction accuracy in a variety of complex situations. This fusion of spatiotemporal features not only enhances the synergy between modal components but also significantly improves the accuracy and effectiveness of prediction results, providing an efficient and highly adaptable solution for accurately predicting the remaining life of lithium batteries.

[0134] Example 2

[0135] See also Figure 2 The second embodiment of the present invention provides a lithium battery life prediction system based on a graph neural network, which is applied to the method described in the above embodiment. The system includes: a data extraction module 10, a sample division module 20, a modal decomposition module 30 and a life prediction module 40.

[0136] A data extraction module 10 is used to obtain charge and discharge cycle data of a lithium battery for several historical periods, and extract battery capacity data based on the charge and discharge cycle data;

[0137] A sample segmentation module 20 is configured to segment the battery capacity data to obtain a plurality of sequence samples corresponding to a time series, and to divide the sequence samples into a training set and a test set according to preset time nodes to train a preset neural network model;

[0138] a modal decomposition module 30 for adaptively adding white noise to the battery capacity data to amplify the degree of non-correlation between modes, thereby decomposing the modes of the battery capacity data;

[0139] The life prediction module 40 is used to construct a graph neural network and combine the adjacency matrix, graph convolution and attention mechanism to obtain a lithium battery life prediction model, and predict the remaining life of the lithium battery through the lithium battery life prediction model.

[0140] Compared to existing technologies, the system shown in this embodiment abandons reliance on manually set parameters and rules and instead employs adaptive learning and feature integration mechanisms, enabling the lithium battery life prediction model to flexibly respond to the variability and complexity of diverse data sets, improving its stability and prediction accuracy in a variety of complex situations. This fusion of spatiotemporal features not only enhances the synergy between modal components but also significantly improves the accuracy and effectiveness of prediction results, providing an efficient and highly adaptable solution for accurately predicting the remaining life of lithium batteries.

[0141] Example 3

[0142] A third embodiment of the present invention provides a readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method described in the above embodiment is implemented.

[0143] Example 4

[0144] A fourth embodiment of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in the above embodiments when executing the computer program.

[0145] Throughout this specification, references to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. Throughout this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0146] The above-described embodiments merely illustrate several implementations of the present invention. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous modifications and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A lithium battery life prediction method based on graph neural network, characterized in that: The method comprises: Acquire charge and discharge cycle data of several historical periods of the lithium battery, and extract battery capacity data based on the charge and discharge cycle data; Segmenting the battery capacity data to obtain a plurality of sequence samples corresponding to a time series, and dividing the sequence samples into a training set and a test set according to preset time nodes to train a preset neural network model; Adaptively adding white noise to the battery capacity data to amplify the degree of non-correlation between modes, so as to decompose the modes of the battery capacity data; specifically comprising: Adding white noise to the original battery capacity data using an integrated empirical mode decomposition method based on adaptive white noise; amplifying the degree of non-correlation between modes in the battery capacity data by adding white noise, so as to perform modal decomposition on the capacity sequence in the battery capacity data; The method further comprises: Construct a graph neural network and combine it with the adjacency matrix, graph convolution and attention mechanism to obtain a lithium battery life prediction model. The remaining life of the lithium battery is predicted by the lithium battery life prediction model. Specifically, it includes: Obtaining the time series; Converting multiple modal components obtained based on capacity sequence decomposition into multiple graph nodes in a modal component graph, calculating the Pearson correlation coefficient between the modal components as elements of an adjacency matrix, and expressing the similarity and dependency relationship between the multiple graph nodes in the modal component graph by using the adjacency matrix; Obtaining the structure of the modal component graph at each moment, employing a graph attention layer to capture the relationship between modal components using an attention mechanism, calculating the attention weights between a modal component and its associated modal components, updating the current modal component state based on the attention weights to extract node spatial features between multiple modal components, and aggregating the node spatial features using average pooling to obtain global spatial features; According to the global spatial features, a gated recurrent unit is used to construct a feature matrix of the time series, and the input time series is modeled through a gating mechanism; The features processed by the graph attention layer and the gated recurrent unit are input into the fully connected layer to predict the remaining life of the lithium battery; the expression is: ； Where, is the predicted value of the remaining life of the lithium battery, is the weight of the fully connected layer, is the hidden state of the last gated recurrent unit, is the bias of the fully connected layer.

2. The lithium battery life prediction method based on graph neural network according to claim 1, characterized in that: The steps of segmenting the battery capacity data to obtain a plurality of sequence samples corresponding to time series, and dividing the sequence samples into a training set and a test set according to preset time nodes to train a preset neural network model include: Continuous battery capacity data is segmented into sliding windows of fixed length. Each fixed-length data segment is used as a sequence sample to capture the dynamic changes in the time series. The target value of each sequence sample is the data of the next time step in the time series, which is used to predict future capacity values. Dividing the plurality of sequence samples according to a preset ratio in chronological order to obtain a training set for training a neural network model and a test set for testing the neural network model; The neural network model is trained using the training set and the test set.

3. The lithium battery life prediction method based on graph neural network according to claim 1, characterized in that: The step of constructing a feature matrix of a time series using a gated recurrent unit according to the global spatial features and modeling the input time series using a gating mechanism comprises: a. Input: Controls the global spatial characteristics of the current input Current status The influence of is expressed as: ； Where, represents the reset gate vector, represents the activation function, and Represent the weight and bias of the gated recurrent unit respectively; b. Update gate: control the previous step state and the global spatial features of the current input The degree of fusion is expressed as: ； Where, To update the gate vector, To update the gate weights, To update the gate bias; c. Candidate state calculation: Generate candidate states at the current moment , the expression is: ； Where, is the candidate state weight, is the candidate state bias; d. State update: Combined update gate vector With candidate status , calculate the final state at the current moment , the expression is: ； e. Every moment , the gated recurrent unit will output a hidden state vector, namely , represents the current moment feature, the last hidden state of the gated recurrent unit It is used as the final feature representation for lithium battery life prediction.

4. The lithium battery life prediction method based on graph neural network according to any one of claims 1 to 3, characterized in that: The method further comprises: The mean absolute percentage error is used as an evaluation index to evaluate the remaining life prediction value output by the lithium battery life prediction model. The expression is: ； Where, is the actual value of the remaining life of the lithium battery, is the predicted value of the remaining life of the lithium battery, is the sample size.

5. A lithium battery life prediction system based on graph neural network, characterized in that: The method according to any one of claims 1 to 4, wherein the system comprises: A data extraction module is used to obtain charge and discharge cycle data of several historical periods of the lithium battery and extract battery capacity data based on the charge and discharge cycle data; a sample segmentation module, configured to segment the battery capacity data to obtain a plurality of sequence samples corresponding to time series, and to divide the sequence samples into a training set and a test set according to preset time nodes, so as to train a preset neural network model; a modal decomposition module, configured to adaptively add white noise to the battery capacity data to amplify the degree of non-correlation between modes, thereby decomposing the modes of the battery capacity data; The life prediction module is used to construct a graph neural network and combine the adjacency matrix, graph convolution and attention mechanism to obtain a lithium battery life prediction model, and predict the remaining life of the lithium battery through the lithium battery life prediction model.

6. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 4 is implemented.

Images