Battery life prediction method and device, electronic equipment and medium

Abstract

The invention provides a battery life prediction method and device, electronic equipment and a medium, and the method comprises the steps: obtaining battery data in a discharging process of a battery and a battery node corresponding to the battery data, and taking the battery data as node feature information corresponding to the battery node; selecting a representative node corresponding to each battery node according to a preset difference threshold; constructing a graph structure according to the node feature information of the battery nodes and the node feature information of representative nodes corresponding to the battery nodes, and processing the graph structure to obtain at least one kind of aggregation feature information of the battery; fusing the at least one kind of aggregation feature information to obtain multi-dimensional spatial-temporal feature information of the battery; and inputting the multi-dimensional spatial-temporal characteristic information into a pre-trained battery life prediction model to obtain a life prediction value of the battery. According to the invention, accurate capture of multi-dimensional spatio-temporal characteristic information in the battery discharge process can be realized, so that the accuracy and real-time performance of battery life prediction are improved.

Description

Technical Field

[0001] The present application relates to the field of vehicle battery technology, and in particular to a battery life prediction method, device, electronic device and medium. Background Art

[0002] As global environmental pollution and climate change become increasingly serious, people's demand for clean energy is increasing day by day. As the main energy source for electric vehicles, lithium-ion batteries have the advantages of high energy density, long life and low pollution, but their performance degradation and potential safety risks need to be paid attention to. Therefore, it is necessary to accurately predict battery life.

[0003] The existing battery life prediction methods are mainly data-driven and neural network-based deep learning methods. However, the data-driven method must calculate the capacity change value under the corresponding voltage, which is a complicated process; the prediction methods based on neural network deep learning methods only consider the time characteristics, ignoring the importance of spatial characteristics for accurate prediction, resulting in limited accuracy of battery life prediction.

[0004] Therefore, how to improve the accuracy of battery life prediction is a problem that needs to be solved at present. Summary of the invention

[0005] In view of the above problems, the embodiments of the present application propose a battery life prediction method, device, electronic device and medium.

[0006] In a first aspect of the implementation of the present application, a battery life prediction method is first provided, the method comprising:

[0007] Acquire battery data during the discharge process of the battery and a battery node corresponding to the battery data, and use the battery data as node feature information corresponding to the battery node;

[0008] Selecting a representative node corresponding to each of the battery nodes according to a preset difference threshold;

[0009] Constructing a graph structure according to the node feature information of the battery node and the node feature information of the representative node corresponding to the battery node, and processing the graph structure to obtain at least one aggregate feature information of the battery;

[0010] fusing at least one of the aggregated feature information to obtain multi-dimensional spatiotemporal feature information of the battery;

[0011] The multi-dimensional spatiotemporal feature information is input into a pre-trained battery life prediction model to obtain a predicted value of the battery life.

[0012] Optionally, selecting a representative node corresponding to each battery node according to a preset difference threshold comprises:

[0013] Taking any node of the battery node as a first node, and acquiring first node characteristic information of the first node;

[0014] Determine the type of the battery data corresponding to the first node;

[0015] Selecting a second node from nodes of the same type as the first node according to a preset battery dimension, and acquiring second node feature information of the second node;

[0016] Calculating a difference between the first node feature information and the second node feature information;

[0017] Inputting the difference into a preset nonlinear activation function to obtain a difference score;

[0018] When the difference score is lower than a preset difference threshold, the second node is a representative node corresponding to the first node.

[0019] Optionally, selecting a second node from nodes of the same type as the first node according to a preset battery dimension includes:

[0020] In the case where the preset battery dimension is the first dimension, selecting a node adjacent to the first node from nodes of the same type as the first node as the second node;

[0021] In the case where the preset battery dimension is the second dimension, selecting a node which is one node away from the first node from nodes of the same type as the first node as the second node;

[0022] In the case where the preset battery dimension is the third dimension, a node which is two nodes away from the first node is selected from nodes of the same type as the first node as the second node.

[0023] Optionally, constructing a graph structure according to the node feature information of the battery node and the node feature information of a representative node corresponding to the battery node includes:

[0024] Calculating a first Euclidean distance between the battery node and the corresponding representative node;

[0025] Combining a number of the first Euclidean distances to obtain a first Euclidean distance matrix;

[0026] The three smallest values ​​in each row of the first Euclidean distance matrix are set as first preset values, and the other values ​​in each row of the first Euclidean distance matrix are set as second preset values, to obtain a first adjacency matrix;

[0027] The characteristic information of the battery node is combined with the first adjacency matrix to construct a first graph structure corresponding to each type of the battery data under each battery dimension.

[0028] Optionally, the processing the graph structure to obtain at least one aggregated feature information of the battery includes:

[0029] Perform graph convolution processing on the first graph structure according to a preset activation function to obtain at least one aggregated feature information of the battery.

[0030] Optionally, fusing at least one of the aggregated feature information to obtain multi-dimensional spatiotemporal feature information of the battery includes:

[0031] Under the same battery dimension, combining the aggregated feature information corresponding to different types of battery data to obtain at least one type of combined feature information corresponding to the battery dimension;

[0032] Performing feature aggregation on at least one of the combined feature information to obtain at least one dimension aggregated feature information corresponding to the battery dimension;

[0033] Performing a maximum pooling operation on at least one of the dimensionally aggregated feature information, and inputting the dimensionally aggregated feature information into a preset neural network fully connected layer and a preset nonlinear activation function to obtain an attention coefficient corresponding to at least one of the battery dimensions;

[0034] Based on at least one of the dimensional aggregated feature information and the corresponding attention coefficient, calculation is performed to obtain the multi-dimensional spatiotemporal feature information of the battery.

[0035] Optionally, the performing feature aggregation on at least one of the combined feature information to obtain at least one dimension aggregated feature information corresponding to the battery dimension includes:

[0036] In the case where the preset battery dimension is the first dimension, a node adjacent to the first node is selected from other nodes as a third node, and the other node is a node different from the first node among the battery nodes;

[0037] When the preset battery dimension is the second dimension, selecting a node which is one node away from the first node from the other nodes as a third node;

[0038] When the preset battery dimension is the third dimension, a node which is two nodes away from the first node is selected from the other nodes as the third node;

[0039] Calculating a second Euclidean distance between the first node and the third node;

[0040] Combining a plurality of the second Euclidean distances to obtain a second Euclidean distance matrix;

[0041] The three smallest values ​​in each row of the second Euclidean distance matrix are set as first preset values, and the other values ​​in each row of the second Euclidean distance matrix are set as second preset values, to obtain a second adjacency matrix;

[0042] Combining the combined feature information with the second adjacency matrix to construct a second graph structure corresponding to each battery dimension;

[0043] The second graph structure is subjected to graph convolution processing according to a preset activation function to obtain at least one dimension aggregation feature information corresponding to the battery dimension.

[0044] In a second aspect of the implementation of the present application, a battery life prediction device is also provided, the device comprising:

[0045] An acquisition module, used to acquire battery data during the discharge process of the battery and a battery node corresponding to the battery data, and use the battery data as node feature information corresponding to the battery node;

[0046] A selection module, used for selecting a representative node corresponding to each of the battery nodes according to a preset difference threshold;

[0047] A construction module, used to construct a graph structure according to the node feature information of the battery node and the node feature information of the representative node corresponding to the battery node, and process the graph structure to obtain at least one aggregate feature information of the battery;

[0048] A fusion module, used to fuse at least one of the aggregated feature information to obtain multi-dimensional spatiotemporal feature information of the battery;

[0049] The prediction module is used to input the multi-dimensional spatiotemporal feature information into a pre-trained battery life prediction model to obtain a predicted value of the battery life.

[0050] Optionally, the selection module includes:

[0051] A first node determination module, used to take any node of the battery node as a first node and obtain first node characteristic information of the first node;

[0052] A type determination module, used to determine the type of the battery data corresponding to the first node;

[0053] A second node determination module, configured to select a second node from nodes of the same type as the first node according to a preset battery dimension, and obtain second node feature information of the second node;

[0054] A first calculation module, used to calculate the difference between the first node characteristic information and the second node characteristic information;

[0055] A difference score determination module, used for inputting the difference into a preset nonlinear activation function to obtain a difference score;

[0056] The representative node determination module is used to determine that the second node is the representative node corresponding to the first node when the difference score is lower than a preset difference threshold.

[0057] Optionally, the second node determination module includes:

[0058] A first selection unit, configured to select, when the preset battery dimension is the first dimension, a node adjacent to the first node from nodes of the same type as the first node as a second node;

[0059] A second selection unit, configured to select, when the preset battery dimension is a second dimension, a node which is one node away from the first node from nodes of the same type as the first node as a second node;

[0060] The third selection unit is used to select a node which is two nodes away from the first node as the second node from the nodes which belong to the same type as the first node when the preset battery dimension is the third dimension.

[0061] Optionally, the building blocks include:

[0062] A first Euclidean distance determination module, used to calculate a first Euclidean distance between the battery node and the corresponding representative node;

[0063] A first Euclidean distance matrix determination module, used for combining a plurality of the first Euclidean distances to obtain a first Euclidean distance matrix;

[0064] A first adjacency matrix determination module, used to set the three smallest values ​​of each row of the first Euclidean distance matrix to first preset values, and set other values ​​of each row of the first Euclidean distance matrix to second preset values, to obtain a first adjacency matrix;

[0065] The first graph structure construction module is used to combine the characteristic information of the battery node with the first adjacency matrix to construct a first graph structure corresponding to each type of the battery data under each battery dimension.

[0066] Optionally, the building block further includes:

[0067] A graph convolution module is used to perform graph convolution processing on the first graph structure according to a preset activation function to obtain at least one aggregate feature information of the battery.

[0068] Optionally, the fusion module includes:

[0069] A combining module, used for combining the aggregated feature information corresponding to different types of battery data under the same battery dimension to obtain at least one type of combined feature information corresponding to the battery dimension;

[0070] A dimension aggregation module, used for performing feature aggregation on at least one of the combined feature information to obtain at least one dimension aggregation feature information corresponding to the battery dimension;

[0071] An attention coefficient determination module, used to perform a maximum pooling operation on at least one of the dimension aggregated feature information, and input the dimension aggregated feature information into a preset neural network fully connected layer and a preset nonlinear activation function to obtain an attention coefficient corresponding to at least one of the battery dimensions;

[0072] The second calculation module is used to calculate and obtain the multi-dimensional spatiotemporal feature information of the battery based on at least one dimension aggregation feature information and the corresponding attention coefficient.

[0073] Optionally, the dimension aggregation module includes:

[0074] a fourth selection unit, configured to select, when the preset battery dimension is the first dimension, a node adjacent to the first node from other nodes as a third node, wherein the other node is a node different from the first node among the battery nodes;

[0075] a fifth selecting unit, configured to select, when the preset battery dimension is the second dimension, a node which is one node away from the first node from the other nodes as a third node;

[0076] a sixth selection unit, configured to select, when the preset battery dimension is a third dimension, a node which is two nodes away from the first node from the other nodes as a third node;

[0077] A second Euclidean distance determining unit, configured to calculate a second Euclidean distance between the first node and the third node;

[0078] A second Euclidean distance matrix determining unit, configured to combine a plurality of said second Euclidean distances to obtain a second Euclidean distance matrix;

[0079] A second adjacency matrix determining unit, configured to set the three smallest values ​​of each row of the second Euclidean distance matrix to first preset values, and set other values ​​of each row of the second Euclidean distance matrix to second preset values, to obtain a second adjacency matrix;

[0080] A second graph structure construction unit, used for combining the combined feature information with the second adjacency matrix to construct a second graph structure corresponding to each battery dimension;

[0081] The first graph convolution unit is used to perform graph convolution processing on the second graph structure according to a preset activation function to obtain at least one dimensional aggregation feature information corresponding to the battery dimension.

[0082] In a third aspect of the implementation of the present application, there is also provided an electronic device, comprising a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program implements the battery life prediction method as described above when executed by the processor.

[0083] In a fourth aspect of the implementation of the present application, a computer-readable storage medium is further provided, on which a computer program is stored. When the computer program is executed by a processor, the battery life prediction method as described above is implemented.

[0084] The embodiments of the present application have the following advantages:

[0085] The embodiment of the present application provides a battery life prediction method, device, electronic device and medium, the method comprising: obtaining battery data during the discharge process of the battery and the battery node corresponding to the battery data, and using the battery data as the node feature information corresponding to the battery node; selecting the representative node corresponding to each battery node according to a preset difference threshold; constructing a graph structure according to the node feature information of the battery node and the node feature information of the representative node corresponding to the battery node, and processing the graph structure to obtain at least one aggregate feature information of the battery; fusing at least one aggregate feature information to obtain multi-dimensional spatiotemporal feature information of the battery; inputting the multi-dimensional spatiotemporal feature information into a pre-trained battery life prediction model to obtain the life prediction value of the battery. Through this application, the multi-dimensional spatiotemporal features during the discharge process of the battery can be accurately captured, thereby improving the accuracy and real-time performance of the battery life prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art.

[0087] Figure 1 is a flowchart of a battery life prediction method provided by an embodiment of the present application;

[0088] Figure 2 It is a structural schematic diagram of a battery life prediction device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0089] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, each embodiment of the present application will be described in detail below in conjunction with the accompanying drawings. However, it will be appreciated by those skilled in the art that in each embodiment of the present application, many technical details are proposed in order to enable the reader to better understand the present application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical scheme claimed in the present application can also be implemented. The division of the following embodiments is for the convenience of description, and the specific implementation of the present application should not constitute any limitation, and the various embodiments can be combined with each other and referenced to each other without contradiction.

[0090] Lithium-ion batteries are highly efficient, lightweight and energy-dense rechargeable batteries that can be widely used in portable electronic devices, electric vehicles and energy storage systems. The positive electrode of a lithium-ion battery is usually made of lithium oxide, such as lithium cobalt oxide, lithium nickel oxide or lithium iron phosphate, while the negative electrode is made of graphite or other materials that can embed lithium ions. The electrolyte is usually a liquid or solid organic solvent that can conduct lithium ions.

[0091] Battery life is the time or number of cycles that a battery can continue to provide effective power under specific conditions. It usually depends on the battery's chemistry, design, usage and environmental conditions. Battery life can be divided into cycle life and calendar life: cycle life refers to the number of times a battery can maintain its performance during charge and discharge cycles; calendar life refers to how long a battery can maintain its performance when stored or idle. As the battery is used, its capacity will gradually decrease until it can no longer meet the needs of the device, at which point the battery reaches the end of its life.

[0092] The data-driven method of predicting battery life can mine the hidden information and evolution laws of battery performance changes from historical data, without the need to accurately model the complex electrochemical mechanisms inside the battery or identify the battery's own parameters. For example, a support vector machine can extract health indicators (HI) from the discharge curve of early cycles to predict battery cycle life, but the HI selected by this method of predicting battery life must calculate the capacity change value at the corresponding voltage, which is a complicated process.

[0093] The capacity of lithium-ion batteries is predicted by convolutional neural networks (CNNs). Due to the local connection and weight sharing characteristics of CNNs, the number of training parameters is small, which alleviates the overfitting problem to a certain extent. However, the characteristics of different time scales of each input parameter of the battery are not considered.

[0094] A battery life prediction method provided by an embodiment of the present application can improve the accuracy of battery life prediction.

[0095] Reference Figure 1 , shows a flowchart of a battery life prediction method provided by an embodiment of the present application, the method may specifically include the following steps:

[0096] Step 101 : acquiring battery data during a discharge process of a battery and a battery node corresponding to the battery data, and using the battery data as node feature information corresponding to the battery node.

[0097] It should be noted that the discharge process of a battery may refer to the process in which the chemical energy inside the battery is converted into electrical energy and released through an external circuit. Battery data may refer to parameters such as the voltage, current, and temperature of the battery that change during the discharge process of the battery. Each battery data corresponds to a battery node, and the characteristic information of the node is the battery data.

[0098] In the embodiment of the present application, sensors and data acquisition systems can be used to monitor the voltage, current, temperature and other parameters of the battery in the discharge process in real time. The collected battery data can be preprocessed, such as filtering, normalization and outlier detection, to ensure the accuracy and consistency of the data. Each battery data corresponds to a node, and the battery data is used as the feature information of the corresponding node.

[0099] Among them, the data acquisition system can monitor the battery's voltage, current, temperature and other parameters in real time by connecting various types of sensors (such as voltage sensors, current sensors, temperature sensors, etc.), and can transmit the collected data to the processing unit for preprocessing and analysis to improve the accuracy and consistency of the data, thereby providing data support for battery life prediction.

[0100] Step 102: Select a representative node corresponding to each battery node according to a preset difference threshold.

[0101] It should be noted that the representative node may be a node in the battery node that is similar to a battery node. The difference threshold may be a threshold of the difference between the battery node and the representative node, which may be used for selecting the representative node.

[0102] In the embodiment of the present application, a representative node corresponding to each battery node may be selected according to a preset difference threshold.

[0103] In an optional embodiment of the present application, step 102 includes the following sub-steps:

[0104] S21: taking any node of the battery node as a first node, and obtaining first node characteristic information of the first node;

[0105] S22: Determine the type of battery data corresponding to the first node;

[0106] S23: selecting a second node from nodes of the same type as the first node according to a preset battery dimension, and acquiring second node feature information of the second node;

[0107] S24: Calculate the difference between the first node feature information and the second node feature information;

[0108] S25: inputting the difference into a preset nonlinear activation function to obtain a difference score;

[0109] S26: When the difference score is lower than a preset difference threshold, the second node is a representative node corresponding to the first node.

[0110] It should be noted that the type of battery data may refer to the type of battery data, such as battery data as voltage or voltage data as current. Battery dimension may refer to the size of the receptive field in a graph convolutional neural network (GCN). Specifically, by setting different dimension values ​​(dimension = 2, 3, 4), receptive fields of different dimensions can be introduced into the graph convolutional neural network, thereby aggregating features within different node hop counts. A nonlinear activation function may be an activation function of a neural network, which can be used to enable a neural network to learn and simulate complex input-output relationships. Commonly used nonlinear activation functions include sigmoid (S-type function).

[0111] Among them, the graph convolutional neural network can be a deep learning model for processing graph structure data. By performing convolution operations on the graph structure, node features and graph structure information can be extracted, which can be applied to social networks, recommendation systems, chemical molecular structure analysis and other fields. The graph structure can be a data structure used to represent the relationship between objects. It consists of nodes and edges, where nodes represent objects and edges represent the relationship between objects. Graph structures are widely used in computer science, mathematics, physics, biology and other fields for modeling and analyzing complex systems. The receptive field can be the size of the area on the input image that can be mapped back to the pixel points on the feature map output by each layer of the graph convolutional neural network, which can represent the area where the graph convolutional neural network features can see the input image.

[0112] In an embodiment of the present application, any node of the battery node can be used as the first node, and the first node characteristic information of the first node can be obtained. The battery data type corresponding to the first node can be determined, such as voltage data, current data, temperature data, etc. The second node can be selected from the nodes of the same type as the first node according to the preset battery dimension (such as the first dimension, the second dimension, and the third dimension), and the characteristic information of the second node can be obtained.

[0113] The difference between the feature information of the first node and the second node can be calculated and input into a preset nonlinear activation function to obtain the difference score of the two nodes. The difference score is compared with a preset difference threshold. When the difference score is lower than the preset difference threshold, it indicates that the second node is a node with a high correlation with the first node, which can reduce the computational complexity while improving the propagation efficiency of the feature information.

[0114] Exemplarily, the difference between the first node feature information and the second node feature information is calculated, and the calculation expression is:

[0115] f 1，j =f1-f j (1)

[0116] Among them, f1 is the feature information of the first node, f j is the feature information of the second node, f 1，j is the difference between the first node feature information and the second node feature information.

[0117] The difference is input into the preset nonlinear activation function to obtain the difference score, and the calculation expression is:

[0118] a 1，j =Sigmoid(f 1，j )(2)

[0119] Among them, a 1，j is the difference score, and the Sigmoid function is the preset activation function.

[0120] When the difference score is lower than a preset difference threshold, the second node is a representative node corresponding to the first node, where the difference threshold is 0.3.

[0121] In the embodiment of the present application, a representative node corresponding to each battery node is selected according to a preset difference threshold, thereby achieving efficient screening of battery nodes, thereby improving the accuracy and efficiency of battery life prediction.

[0122] In an optional embodiment of the present application, step S23 includes the following sub-steps:

[0123] S231: when the preset battery dimension is the first dimension, selecting a node adjacent to the first node from nodes of the same type as the first node as a second node;

[0124] S232: when the preset battery dimension is the second dimension, selecting a node which is one node away from the first node from nodes of the same type as the first node as the second node;

[0125] S233: When the preset battery dimension is the third dimension, a node which is two nodes away from the first node is selected from nodes of the same type as the first node as the second node.

[0126] In an embodiment of the present application, when the preset battery dimension is the first dimension, a node that reaches the first node through 1 hop is selected from the nodes of the same type as the first node as the second node; when the preset battery dimension is the second dimension, a node that reaches the first node through 2 hops is selected from the nodes of the same type as the first node as the second node; when the preset battery dimension is the third dimension, a node that reaches the first node through 3 hops is selected from the nodes of the same type as the first node as the second node.

[0127] In an embodiment of the present application, a second node at a different distance from the first node is selected according to different battery dimensions, and the correlation between battery nodes can be analyzed from a multi-dimensional perspective, thereby more comprehensively capturing the spatiotemporal characteristics of battery data.

[0128] Step 103: construct a graph structure according to the node feature information of the battery node and the node feature information of the representative node corresponding to the battery node, and process the graph structure to obtain at least one aggregate feature information of the battery.

[0129] It should be noted that the aggregated feature information may be feature information obtained by aggregating different types of battery data at different battery dimensions.

[0130] In an embodiment of the present application, the node feature information of the battery node and the node feature information of the corresponding representative node can be used as a feature information set of the graph structure. The relationship between the battery node and the representative node can be used as an edge of the graph structure. The two constitute a graph structure, and a graph convolutional neural network is used to perform a convolution operation on the constructed graph structure, aggregate the feature information of the neighborhood nodes, and obtain at least one aggregate feature information.

[0131] In an optional embodiment of the present application, step 103 includes the following sub-steps:

[0132] S31: Calculate the first Euclidean distance between the battery node and the corresponding representative node;

[0133] S32: Combining a plurality of first Euclidean distances to obtain a first Euclidean distance matrix;

[0134] S33: setting the three smallest values ​​in each row of the first Euclidean distance matrix as first preset values, and setting the other values ​​in each row of the first Euclidean distance matrix as second preset values, to obtain a first adjacency matrix;

[0135] S34: Combine the characteristic information of the battery node with the first adjacency matrix to construct a first graph structure corresponding to each type of battery data in each battery dimension.

[0136] S35: Performing graph convolution processing on the first graph structure according to a preset activation function to obtain at least one aggregated feature information of the battery.

[0137] It should be noted that the Euclidean distance can be a distance measurement method used to measure the straight-line distance between two points in a multidimensional space. The first preset value is 1, and the second preset value is 0, 1 represents that the nodes are connected, and 0 represents that the nodes are not connected. The adjacency matrix can be a data structure used to represent the graph structure. Its rows and columns correspond to the nodes in the graph, and the elements in the matrix represent the connection relationship between the nodes. The adjacency matrix can be applied to graph theory, network analysis, machine learning and other fields. The graph structure can be represented by G(X, A), where X is all the node feature information, and A is the adjacency matrix, representing the connection relationship between the nodes.

[0138] In an embodiment of the present application, the Euclidean distance between the battery node and the corresponding representative node can be calculated, and the obtained Euclidean distances can be combined into a Euclidean distance matrix. The points corresponding to the three smallest Euclidean distances in each row of the distance matrix can be set to 1, and the remaining points in each row can be set to 0, thereby obtaining an adjacency matrix representing the connection relationship between the nodes. The node feature information and the adjacency matrix are combined to obtain a graph structure, each graph structure corresponding to a battery dimension and a type of battery data. For example, if the representative node is selected under the first dimension, and the battery node is a node corresponding to the current data, the graph structure calculated at this time is the graph structure corresponding to the current data under the first dimension. The graph structure can be subjected to graph convolution processing to obtain aggregate feature information corresponding to each type of battery data under each dimension.

[0139] Exemplarily, the first graph structure is subjected to graph convolution processing according to a preset activation function to obtain at least one aggregate feature information of the battery, and the calculation expression is:

[0140]

[0141] Among them, k represents the battery dimension, i represents the battery data type, Represents the adjacency matrix Perform symmetric normalization operation, is the node feature information of the battery node and the representative node in the i-th battery data type under the k-th battery dimension, is the weight matrix, ReLU (Rectified Linear Unit) is the preset activation function, It is the aggregate feature information corresponding to the i-th battery data type under the k-th battery dimension.

[0142] In an embodiment of the present application, a graph structure is constructed for each type of data in each dimension, so as to aggregate neighborhood features within different receptive fields and effectively capture the complex relationships between battery nodes.

[0143] Step 104: fuse at least one type of aggregated feature information to obtain multi-dimensional spatiotemporal feature information of the battery.

[0144] In the embodiment of the present application, aggregate feature information of different dimensions and types can be fused to obtain multi-dimensional spatiotemporal features of the battery.

[0145] In an optional embodiment of the present application, step 104 includes the following sub-steps:

[0146] S41: Under the same battery dimension, aggregate feature information corresponding to different types of battery data is combined to obtain at least one type of combined feature information corresponding to the battery dimension;

[0147] S42: performing feature aggregation on at least one type of combined feature information to obtain at least one type of dimension aggregated feature information corresponding to a battery dimension;

[0148] S43: performing a maximum pooling operation on at least one dimension aggregated feature information, and inputting the dimension aggregated feature information into a preset neural network fully connected layer and a preset nonlinear activation function to obtain an attention coefficient corresponding to at least one battery dimension;

[0149] S44: Calculate and obtain multi-dimensional spatiotemporal feature information of the battery based on at least one dimension of aggregated feature information and a corresponding attention coefficient.

[0150] It should be noted that the maximum pooling operation is a downsampling method that can be used to reduce the size of the feature map while retaining the most important feature information. It can be applied in image processing and computer vision tasks. The fully connected layer of a neural network is a basic layer type in a neural network. It can connect all neurons in the previous layer with all neurons in the current layer to learn complex nonlinear relationships. The attention coefficient can be used to assign different weights to different parts of the neural network input data.

[0151] In an embodiment of the present application, the aggregated feature information corresponding to different types of battery data under the same battery dimension can be combined to obtain the combined features corresponding to each battery dimension. Feature aggregation is performed on each combined feature information to obtain the dimensional aggregated feature information corresponding to each battery dimension. The maximum pooling operation is performed on the dimensional aggregated feature information to extract the most significant node feature information of all types of battery data in each dimension, and then the most significant node feature information is input into the fully connected layer of the neural network, and the attention coefficient corresponding to each battery dimension is obtained through a nonlinear activation function. The multi-dimensional spatiotemporal feature information of the battery is calculated based on the combined features and attention coefficients corresponding to each battery dimension.

[0152] Exemplarily, according to the three types of battery dimension aggregated feature information and the corresponding attention coefficients, the multi-dimensional spatiotemporal feature information of the battery is calculated and the calculation expression is:

[0153] H out =a1H d1 +a2H d2 +a3H d3 (4)

[0154] Among them, H out is the multi-dimensional spatiotemporal feature information, a1 is the attention coefficient corresponding to the first battery dimension, a2 is the attention coefficient corresponding to the second battery dimension, a3 is the attention coefficient corresponding to the third battery dimension, H d1 is the dimension aggregation feature information corresponding to the first battery dimension, Hd2 is the dimension aggregation feature information corresponding to the second battery dimension, H d3 Aggregate feature information for the dimension corresponding to the third battery dimension.

[0155] In the embodiments of the present application, aggregated feature information of different dimensions and types is effectively integrated, thereby improving the extraction accuracy and calculation efficiency of multi-dimensional spatiotemporal feature information of the battery, thereby improving the accuracy of battery life prediction.

[0156] In an optional embodiment of the present application, step S42 further includes the following sub-steps:

[0157] S421: When the preset battery dimension is the first dimension, select a node adjacent to the first node from other nodes as a third node, and the other node is a node different from the first node among the battery nodes;

[0158] S422: When the preset battery dimension is the second dimension, select a node that is one node away from the first node from other nodes as a third node;

[0159] S423: When the preset battery dimension is the third dimension, select a node that is two nodes away from the first node from other nodes as the third node;

[0160] S424: Calculate the second Euclidean distance between the first node and the third node;

[0161] S425: combining a plurality of second Euclidean distances to obtain a second Euclidean distance matrix;

[0162] S426: setting the three smallest values ​​in each row of the second Euclidean distance matrix to first preset values, and setting the other values ​​in each row of the second Euclidean distance matrix to second preset values, to obtain a second adjacency matrix;

[0163] S427: Combining the combined feature information with the second adjacency matrix to construct a second graph structure corresponding to each battery dimension;

[0164] S428: Perform graph convolution processing on the second graph structure according to a preset activation function to obtain at least one dimension aggregation feature information corresponding to the battery dimension.

[0165] It should be noted that the other nodes may be other nodes among all battery nodes except the first node. Selecting the third node from other nodes can be used to calculate the connection relationship between all battery nodes, not limited to the connection relationship between battery nodes of the same battery data type.

[0166] In an embodiment of the present application, when the preset battery dimension is the first dimension, a node that reaches the first node through 1 hop is selected from other battery nodes as the third node; when the preset battery dimension is the second dimension, a node that reaches the first node through 2 hops is selected from other nodes as the third node; when the preset battery dimension is the third dimension, a node that reaches the first node through 3 hops is selected from other nodes as the third node. The Euclidean distance between the first node and the third node can be calculated, and the obtained Euclidean distance is combined into a Euclidean distance matrix to quantify the similarity between the nodes and systematically represent the distance relationship between the nodes. The points corresponding to the three smallest Euclidean distances in each row of the Euclidean distance matrix can be set to 1, and the remaining nodes in each row can be set to 0, so as to obtain an adjacency matrix representing the connection relationship between the nodes, which can simplify the connection relationship between the nodes and highlight the important connection relationship. The combined feature information is combined with the second adjacency matrix to construct a graph structure corresponding to each battery dimension, and the node features and the connection relationship between the nodes can be combined to form a complete graph representation. The graph structure can be processed by graph convolution to obtain the corresponding dimension aggregation feature information under each battery dimension.

[0167] Exemplarily, the second graph structure is subjected to graph convolution processing according to a preset activation function to obtain at least one dimension aggregation feature information corresponding to the battery dimension, and the calculation expression is:

[0168]

[0169] Where k represents the battery dimension, Represents the adjacency matrix A k Perform symmetric normalization operation, X k is the combined feature information corresponding to the kth battery dimension, W k is the weight matrix, ReLU is the preset activation function, H k Aggregate feature information for the corresponding dimension under the k-th battery dimension.

[0170] In the embodiment of the present application, the spatial feature relationship between various battery data in the discharge process under the same dimension can be mined to improve the accuracy of battery life prediction.

[0171] Step 105, input the multi-dimensional spatiotemporal feature information into a pre-trained battery life prediction model to obtain a battery life prediction value.

[0172] It should be noted that the pre-trained battery life prediction model can be a machine learning or deep learning model that has been trained with a large amount of battery data. The model can learn the complex relationship between battery data and battery life. The battery life prediction value can be an estimate of the remaining battery life, the time point when the battery reaches the end of its life, or the number of times the battery can be charged.

[0173] In the embodiment of the present application, multi-dimensional spatiotemporal feature information can be used as input to a pre-trained battery life prediction model. The feature information contains the complex behavior and state of the battery in different dimensions and time, which can provide rich contextual information for the model, thereby improving the accuracy of the prediction. The model can perform inference calculations based on its internal learning parameters and structure, and output the predicted value of the battery life.

[0174] In an optional embodiment of the present application, when training a battery life prediction model, a mean square error can be used as a loss function to measure the difference between the predicted value and the true value. The model can be trained through multiple iterations, and the parameters can be continuously adjusted to improve the model performance. The trained model is deployed in a production environment for prediction of actual battery life.

[0175] The embodiment of the present application provides a battery life prediction method, device, electronic device and medium, the method comprising: obtaining battery data during the discharge process of the battery and the battery node corresponding to the battery data, and using the battery data as the node feature information corresponding to the battery node; selecting the representative node corresponding to each battery node according to a preset difference threshold; constructing a graph structure according to the node feature information of the battery node and the node feature information of the representative node corresponding to the battery node, and processing the graph structure to obtain at least one aggregate feature information of the battery; fusing at least one aggregate feature information to obtain multi-dimensional spatiotemporal feature information of the battery; inputting the multi-dimensional spatiotemporal feature information into a pre-trained battery life prediction model to obtain the life prediction value of the battery. Through this application, the multi-dimensional spatiotemporal features during the discharge process of the battery can be accurately captured, thereby improving the accuracy and real-time performance of the battery life prediction.

[0176] Reference Figure 2 , shows a schematic diagram of the structure of a battery life prediction device provided by an embodiment of the present application, which may specifically include the following modules:

[0177] An acquisition module 201 is used to acquire battery data during the discharge process of the battery and a battery node corresponding to the battery data, and use the battery data as node feature information corresponding to the battery node;

[0178] A selection module 202 is used to select a representative node corresponding to each battery node according to a preset difference threshold;

[0179] A construction module 203 is used to construct a graph structure according to the node feature information of the battery node and the node feature information of the representative node corresponding to the battery node, and process the graph structure to obtain at least one aggregate feature information of the battery;

[0180] A fusion module 204, used to fuse at least one type of aggregated feature information to obtain multi-dimensional spatiotemporal feature information of the battery;

[0181] The prediction module 205 is used to input the multi-dimensional spatiotemporal feature information into the pre-trained battery life prediction model to obtain a battery life prediction value.

[0182] In an optional embodiment of the present application, the selection module 202 includes:

[0183] A first node determination module, used to take any node of the battery node as a first node and obtain first node characteristic information of the first node;

[0184] A type determination module, used to determine the type of battery data corresponding to the first node;

[0185] A second node determination module, configured to select a second node from nodes of the same type as the first node according to a preset battery dimension, and obtain second node feature information of the second node;

[0186] A first calculation module, used to calculate the difference between the first node characteristic information and the second node characteristic information;

[0187] A difference score determination module is used to input the difference value into a preset nonlinear activation function to obtain a difference score;

[0188] The representative node determination module is used to determine that the second node is the representative node corresponding to the first node when the difference score is lower than a preset difference threshold.

[0189] In an optional embodiment of the present application, the second node determination module includes:

[0190] A first selection unit, configured to select, when the preset battery dimension is the first dimension, a node adjacent to the first node from nodes of the same type as the first node as a second node;

[0191] A second selection unit, configured to select, when the preset battery dimension is the second dimension, a node which is one node away from the first node from nodes of the same type as the first node as the second node;

[0192] The third selection unit is used to select a node which is two nodes away from the first node as the second node from nodes which belong to the same type as the first node when the preset battery dimension is the third dimension.

[0193] In an optional embodiment of the present application, the construction module 203 includes:

[0194] A first Euclidean distance determination module, used to calculate a first Euclidean distance between a battery node and a corresponding representative node;

[0195] A first Euclidean distance matrix determination module, used for combining a plurality of first Euclidean distances to obtain a first Euclidean distance matrix;

[0196] A first adjacency matrix determination module, used to set the three smallest values ​​of each row of the first Euclidean distance matrix to first preset values, and set other values ​​of each row of the first Euclidean distance matrix to second preset values, to obtain a first adjacency matrix;

[0197] The first graph structure building module is used to combine the characteristic information of the battery node with the first adjacency matrix to build a first graph structure corresponding to each type of battery data in each battery dimension.

[0198] In an optional embodiment of the present application, the construction module 203 further includes:

[0199] The graph convolution module is used to perform graph convolution processing on the first graph structure according to a preset activation function to obtain at least one aggregated feature information of the battery.

[0200] In an optional embodiment of the present application, the fusion module 204 includes:

[0201] A combination module, used to combine the aggregated feature information corresponding to different types of battery data under the same battery dimension to obtain at least one type of combined feature information corresponding to the battery dimension;

[0202] A dimension aggregation module, used to perform feature aggregation on at least one combination of feature information to obtain at least one dimension aggregation feature information corresponding to a battery dimension;

[0203] An attention coefficient determination module is used to perform a maximum pooling operation on at least one dimension aggregated feature information, and input the dimension aggregated feature information into a preset neural network fully connected layer and a preset nonlinear activation function to obtain an attention coefficient corresponding to at least one battery dimension;

[0204] The second calculation module is used to calculate and obtain multi-dimensional spatiotemporal feature information of the battery based on at least one dimension of aggregated feature information and a corresponding attention coefficient.

[0205] In an optional embodiment of the present application, the dimension aggregation module includes:

[0206] a fourth selection unit, configured to select, when the preset battery dimension is the first dimension, a node adjacent to the first node from other nodes as a third node, wherein the other nodes are nodes different from the first node in the battery nodes;

[0207] a fifth selection unit, configured to select, when the preset battery dimension is the second dimension, a node which is one node away from the first node from other nodes as a third node;

[0208] a sixth selection unit, configured to select, when the preset battery dimension is the third dimension, a node which is two nodes away from the first node from other nodes as the third node;

[0209] A second Euclidean distance determination unit, used to calculate a second Euclidean distance between the first node and the third node;

[0210] A second Euclidean distance matrix determination unit, used for combining a plurality of second Euclidean distances to obtain a second Euclidean distance matrix;

[0211] A second adjacency matrix determining unit, used to set the three smallest values ​​of each row of the second Euclidean distance matrix to first preset values, and set other values ​​of each row of the second Euclidean distance matrix to second preset values, to obtain a second adjacency matrix;

[0212] A second graph structure construction unit, used to combine the combined feature information with the second adjacency matrix to construct a second graph structure corresponding to each battery dimension;

[0213] The first graph convolution unit is used to perform graph convolution processing on the second graph structure according to a preset activation function to obtain at least one dimension aggregation feature information corresponding to the battery dimension.

[0214] An embodiment of the present application also provides an electronic device, which may include a processor, a memory, and a computer program stored in the memory and capable of running on the processor. When the computer program is executed by the processor, the battery life prediction method as described above is implemented.

[0215] An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the battery life prediction method as described above is implemented.

[0216] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0217] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application 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 the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0218] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0219] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, devices, or computer program products. Therefore, the embodiments of the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the embodiments of the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0220] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0221] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0222] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0223] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present application.

[0224] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "including one..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the above elements.

[0225] The battery life prediction method, device, equipment, and medium provided above are introduced in detail. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A battery life prediction method, characterized in that: The method comprises: Acquire battery data during the discharge process of the battery and a battery node corresponding to the battery data, and use the battery data as node feature information corresponding to the battery node; Selecting a representative node corresponding to each of the battery nodes according to a preset difference threshold; Constructing a graph structure according to the node feature information of the battery node and the node feature information of the representative node corresponding to the battery node, and processing the graph structure to obtain at least one aggregate feature information of the battery; fusing at least one of the aggregated feature information to obtain multi-dimensional spatiotemporal feature information of the battery; The multi-dimensional spatiotemporal feature information is input into a pre-trained battery life prediction model to obtain a predicted value of the battery life.

2. The method according to claim 1, characterized in that The selecting a representative node corresponding to each battery node according to a preset difference threshold comprises: Taking any node of the battery node as a first node, and acquiring first node characteristic information of the first node; Determine the type of the battery data corresponding to the first node; Selecting a second node from nodes of the same type as the first node according to a preset battery dimension, and acquiring second node feature information of the second node; Calculating a difference between the first node feature information and the second node feature information; Inputting the difference into a preset nonlinear activation function to obtain a difference score; When the difference score is lower than a preset difference threshold, the second node is a representative node corresponding to the first node.

3. The method according to claim 2, characterized in that The selecting a second node from nodes of the same type as the first node according to a preset battery dimension includes: In the case where the preset battery dimension is the first dimension, selecting a node adjacent to the first node from nodes of the same type as the first node as the second node; In the case where the preset battery dimension is the second dimension, selecting a node which is one node away from the first node from nodes of the same type as the first node as the second node; In the case where the preset battery dimension is the third dimension, a node which is two nodes away from the first node is selected from nodes of the same type as the first node as the second node.

4. The method according to claim 2, characterized in that: The constructing a graph structure according to the node feature information of the battery node and the node feature information of the representative node corresponding to the battery node includes: Calculating a first Euclidean distance between the battery node and the corresponding representative node; Combining a number of the first Euclidean distances to obtain a first Euclidean distance matrix; The three smallest values ​​in each row of the first Euclidean distance matrix are set as first preset values, and the other values ​​in each row of the first Euclidean distance matrix are set as second preset values, to obtain a first adjacency matrix; The characteristic information of the battery node is combined with the first adjacency matrix to construct a first graph structure corresponding to each type of the battery data under each battery dimension.

5. The method according to claim 4, characterized in that The processing of the graph structure to obtain at least one aggregated feature information of the battery includes: Perform graph convolution processing on the first graph structure according to a preset activation function to obtain at least one aggregated feature information of the battery.

6. The method according to claim 2, characterized in that The fusing at least one of the aggregated feature information to obtain multi-dimensional spatiotemporal feature information of the battery includes: Under the same battery dimension, combining the aggregated feature information corresponding to different types of battery data to obtain at least one type of combined feature information corresponding to the battery dimension; Performing feature aggregation on at least one of the combined feature information to obtain at least one dimension aggregated feature information corresponding to the battery dimension; Performing a maximum pooling operation on at least one of the dimensionally aggregated feature information, and inputting the dimensionally aggregated feature information into a preset neural network fully connected layer and a preset nonlinear activation function to obtain an attention coefficient corresponding to at least one of the battery dimensions; Based on at least one of the dimensional aggregated feature information and the corresponding attention coefficient, calculation is performed to obtain the multi-dimensional spatiotemporal feature information of the battery.

7. The method according to claim 6, characterized in that The performing feature aggregation on at least one of the combined feature information to obtain at least one dimension aggregated feature information corresponding to the battery dimension includes: In the case where the preset battery dimension is the first dimension, a node adjacent to the first node is selected from other nodes as a third node, and the other node is a node different from the first node among the battery nodes; When the preset battery dimension is the second dimension, selecting a node which is one node away from the first node from the other nodes as a third node; When the preset battery dimension is the third dimension, a node which is two nodes away from the first node is selected from the other nodes as the third node; Calculating a second Euclidean distance between the first node and the third node; Combining a plurality of the second Euclidean distances to obtain a second Euclidean distance matrix; The three smallest values ​​in each row of the second Euclidean distance matrix are set as first preset values, and the other values ​​in each row of the second Euclidean distance matrix are set as second preset values, to obtain a second adjacency matrix; Combining the combined feature information with the second adjacency matrix to construct a second graph structure corresponding to each battery dimension; The second graph structure is subjected to graph convolution processing according to a preset activation function to obtain at least one dimension aggregation feature information corresponding to the battery dimension.

8. A battery life prediction device, characterized in that: The device comprises: An acquisition module, used to acquire battery data during the discharge process of the battery and a battery node corresponding to the battery data, and use the battery data as node feature information corresponding to the battery node; A selection module, used for selecting a representative node corresponding to each of the battery nodes according to a preset difference threshold; A construction module, used to construct a graph structure according to the node feature information of the battery node and the node feature information of the representative node corresponding to the battery node, and process the graph structure to obtain at least one aggregate feature information of the battery; A fusion module, used to fuse at least one of the aggregated feature information to obtain multi-dimensional spatiotemporal feature information of the battery; The prediction module is used to input the multi-dimensional spatiotemporal feature information into a pre-trained battery life prediction model to obtain a predicted value of the battery life.

9. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor is used to implement the battery life prediction method as described in any one of claims 1 to 7 when executing the program stored in the memory.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the battery life prediction method according to any one of claims 1 to 7 is implemented.

Images