Vehicle power distribution method, device and equipment, storage medium and vehicle

By constructing a key parameter diagram structure and a battery life prediction model, combining the minimum fuel consumption strategy, optimizing vehicle power distribution, the problem of under-taking battery health status is solved, and battery life is extended and vehicle performance is improved.

CN119078787APending Publication Date: 2024-12-06CHONGQING SELIS PHOENIX INTELLIGENT INNOVATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411080174.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-07
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The existing vehicle power distribution strategy fails to fully consider the health status and remaining life of the battery, affecting vehicle performance and user costs.

Method used

By constructing a key parameter graph structure, a battery life prediction model is generated based on the association between the key characteristic parameters of the battery and the number of cycles, and combining the minimum fuel consumption strategy, the power distribution parameters are optimized to balance fuel consumption and battery life.

Benefits of technology

It has achieved optimization of vehicle power distribution, extended battery life, reduced user secondary costs, improved vehicle performance and marketization process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119078787A_ABST
    Figure CN119078787A_ABST
Patent Text Reader

Abstract

The invention provides a vehicle power distribution method and device, equipment, a storage medium and a vehicle. The vehicle power distribution method comprises the steps that a key parameter graph structure is constructed according to the incidence relation between key characteristic parameters corresponding to a target battery of a target vehicle and the cycle index; generating a battery life prediction value according to a pre-constructed battery life prediction model and the key parameter graph structure; determining a target power distribution parameter according to the battery life prediction value and a pre-obtained minimum fuel consumption strategy; and controlling the target vehicle to perform power distribution according to the power distribution parameters. According to the embodiment of the invention, the key characteristic parameters related to the service life of the battery are selected, the key parameter graph structure is constructed based on the key characteristic parameters, the optimal target power distribution parameter is determined according to the combination of the constructed battery service life prediction model and the minimum fuel consumption strategy, and then the optimal target power distribution parameter is obtained according to the power distribution scheme. Meanwhile, limitation of the service life of the battery is considered, and balance between optimal fuel consumption and the service life of the battery is achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present application relate to the field of automobile application technology, and in particular, to a vehicle power distribution method, device, equipment, storage medium and vehicle. Background Art

[0002] The extended-range electric vehicle has a simple structure and is easy to control. While reducing the capacity of the power battery, the range extender ensures the vehicle's driving range. It reduces the cost of the entire vehicle while meeting the user's driving needs, making it an ideal model for transitioning to pure electric vehicles. However, looking at the current pure electric vehicles, the life of their power batteries is generally lower than the estimates given by the car manufacturers, and the cost of power batteries is a large part of the cost of extended-range vehicles. The decline in battery life not only affects the vehicle's power performance, but also increases the user's secondary cost consumption. Therefore, based on the consideration of battery life, researching, developing and applying vehicle control strategies has great practical significance for improving the performance and marketization process of extended-range electric vehicles.

[0003] Current vehicle power distribution strategies often do not fully consider the integration of data-driven battery life predictions, resulting in a failure to fully consider the battery's health status and remaining life during the energy management process, thus affecting vehicle performance. Summary of the invention

[0004] Embodiments of the present application provide a vehicle power distribution method, device, system, electronic device, and computer-readable storage medium to solve the problem in related technologies that vehicle power distribution fails to fully consider the health status and remaining life of the battery, thereby affecting vehicle performance.

[0005] In a first aspect, an embodiment of the present application provides a vehicle power distribution method, the method comprising:

[0006] Constructing a key parameter graph structure according to the correlation between the key characteristic parameters corresponding to the target battery of the target vehicle and the number of cycles, wherein the key parameter graph structure includes an adjacency matrix and node eigenvalues, the adjacency matrix is ​​generated based on the multiplication of an initial characteristic matrix of each of the key characteristic parameters and a transposed matrix corresponding to the initial characteristic matrix, and the node eigenvalues ​​include the key characteristic parameters;

[0007] generating a battery life prediction value according to a pre-constructed battery life prediction model and the key parameter graph structure, wherein the pre-constructed battery life prediction model is constructed based on the key parameter graph structure and the target enhancement feature;

[0008] The target vehicle is controlled to distribute power according to the battery life prediction value.

[0009] Optionally, before the step of generating a battery life prediction value according to the pre-built battery life prediction model and the key parameter graph structure, the method includes:

[0010] Extracting spatial features from the key parameter graph structure to obtain initial spatial features;

[0011] Sorting the adjacency matrix to obtain a target adjacency matrix;

[0012] Generating a first key neighborhood graph according to the initial spatial features and the target adjacency matrix;

[0013] Generating a high-order graph structure according to the first key neighborhood graph, wherein the high-order graph structure represents the target space features corresponding to the key feature parameters;

[0014] Performing key neighborhood extraction on the high-order graph structure to obtain a second key neighborhood graph, wherein the second key neighborhood graph includes target space features corresponding to key neighborhood graph nodes extracted from the high-order graph structure;

[0015] Performing enhancement processing on the second spatial feature according to a pre-constructed enhancement feature formula to obtain a target enhanced feature;

[0016] A battery life prediction model is generated according to the target enhanced features.

[0017] Optionally, generating a battery life prediction value according to a pre-built battery life prediction model and the key parameter graph structure includes:

[0018] The target enhanced features are passed through a fully connected layer to output a battery life prediction value.

[0019] Optionally, the key characteristic parameters include at least one of constant current charging time, constant voltage charging time, and internal resistance.

[0020] Optionally, constructing a key parameter graph structure according to the correlation between the key characteristic parameters corresponding to the target battery of the target vehicle and the number of cycles includes:

[0021] Generating a constant current charging time variation curve according to the constant current charging time and the number of cycles, generating a constant voltage charging time variation curve according to the constant voltage charging time and the number of cycles, and generating an internal resistance variation curve according to the internal resistance and the number of cycles;

[0022] According to the constant current charging time variation curve, the constant voltage charging time variation curve and the internal resistance variation curve, the number of cycles is divided according to preset window values ​​and features are extracted to obtain an initial feature matrix;

[0023] Generate an adjacency matrix based on the multiplication of an initial feature matrix of each of the key feature parameters and a transposed matrix corresponding to the initial feature matrix;

[0024] A key parameter graph structure is constructed according to the initial feature matrix and the adjacency matrix.

[0025] Optionally, determining the target power distribution parameter according to the battery life prediction value and the pre-acquired minimum fuel consumption strategy includes:

[0026] Determine an equivalent fuel consumption value according to a pre-acquired minimum fuel consumption strategy, and use the equivalent fuel consumption value as a first objective function and the battery life prediction value as a second objective function;

[0027] The target power distribution parameter is determined by a preset genetic algorithm, the first objective function and the second objective function.

[0028] In a second aspect, an embodiment of the present application provides a vehicle power distribution device, the device comprising:

[0029] A key parameter graph structure construction module, used to construct a key parameter graph structure according to the correlation between the key characteristic parameters corresponding to the target battery of the target vehicle and the number of cycles, wherein the key parameter graph structure includes an adjacency matrix and node eigenvalues, the adjacency matrix is ​​generated based on the multiplication of an initial characteristic matrix of each of the key characteristic parameters and a transposed matrix corresponding to the initial characteristic matrix, and the node eigenvalues ​​include the key characteristic parameters;

[0030] A generating module, configured to generate a battery life prediction value according to a pre-constructed battery life prediction model and the key parameter graph structure, wherein the pre-constructed battery life prediction model is constructed based on the key parameter graph structure and the target enhancement feature;

[0031] A distribution module is used to control the target vehicle to distribute power according to the battery life prediction value.

[0032] Optionally, the device further comprises:

[0033] The first module is used to extract spatial features from the key parameter graph structure to obtain initial spatial features;

[0034] The second module is used to sort the adjacency matrix to obtain a target adjacency matrix;

[0035] A third module is used to generate a first key neighborhood graph according to the initial spatial features and the target adjacency matrix;

[0036] A fourth module is used to generate a high-order graph structure according to the first key neighborhood graph, wherein the high-order graph structure represents the target space features corresponding to the key feature parameters;

[0037] A fifth module is used to extract key neighborhoods for the high-order graph structure to obtain a second key neighborhood graph, wherein the second key neighborhood graph includes target space features corresponding to key neighborhood graph nodes extracted from the high-order graph structure;

[0038] A sixth module is used to enhance the second spatial feature according to a pre-constructed enhancement feature formula to obtain a target enhancement feature;

[0039] The seventh module is used to generate a battery life prediction model according to the target enhanced features.

[0040] Optionally, the generating module includes:

[0041] The generation submodule is used to output the battery life prediction value through a fully connected layer through the target enhanced features.

[0042] Optionally, the key parameter graph structure building module includes:

[0043] A first construction submodule is used to generate a constant current charging time variation curve according to the constant current charging time and the number of cycles, to generate a constant voltage charging time variation curve according to the constant voltage charging time and the number of cycles, and to generate an internal resistance variation curve according to the internal resistance and the number of cycles;

[0044] A second construction submodule is used to divide the number of cycles according to the constant current charging time variation curve, the constant voltage charging time variation curve and the internal resistance variation curve according to preset window values ​​and perform feature extraction to obtain an initial feature matrix;

[0045] A third construction submodule is used to generate an adjacency matrix based on multiplying an initial feature matrix of each of the key feature parameters by a transposed matrix corresponding to the initial feature matrix;

[0046] The fourth construction submodule is used to construct a key parameter graph structure according to the initial feature matrix and the adjacency matrix.

[0047] Optionally, the determining module includes:

[0048] A first determination submodule, configured to determine an equivalent fuel consumption value according to a pre-acquired minimum fuel consumption strategy, and use the equivalent fuel consumption value as a first objective function and use the battery life prediction value as a second objective function;

[0049] The second determination submodule is used to determine the target power distribution parameter by using a preset genetic algorithm, the first objective function and the second objective function.

[0050] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: a processor; and a memory for storing instructions executable by the processor, wherein the processor is configured to execute the instructions to implement any of the vehicle power distribution methods described above.

[0051] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which, when instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to execute any of the vehicle power distribution methods described above.

[0052] In a fifth aspect, an embodiment of the present application further provides a vehicle, comprising the vehicle power distribution device as described above.

[0053] In an embodiment of the present application, a key parameter graph structure is constructed according to the correlation between the key characteristic parameters corresponding to the target battery of the target vehicle and the number of cycles, wherein the key parameter graph structure includes an adjacency matrix and node eigenvalues, wherein the adjacency matrix is ​​generated based on the multiplication of the initial characteristic matrix of each of the key characteristic parameters and the transposed matrix corresponding to the initial characteristic matrix, and the node eigenvalues ​​include the key characteristic parameters; a battery life prediction value is generated according to a pre-constructed battery life prediction model and the key parameter graph structure, wherein the pre-constructed battery life prediction model is constructed based on the key parameter graph structure and target enhancement features; a target power distribution parameter is determined according to the battery life prediction value and a pre-acquired minimum fuel consumption strategy; and the target vehicle is controlled to distribute power according to the power distribution parameter. That is, the embodiment of the present application selects key characteristic parameters related to battery life, constructs a key parameter graph structure based on the key characteristic parameters, and then combines the constructed battery life prediction model with the minimum fuel consumption strategy, and determines the optimal target power distribution parameter by continuously evolving and optimizing the power distribution parameter, and then achieves the best balance between fuel consumption and battery life according to the power distribution scheme and the limitation of battery life while considering the battery life.

[0054] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present application. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0056] Figure 1 is a flow chart of the steps of a vehicle power distribution method provided by an embodiment of the present application;

[0057] Figure 2 is a flowchart of another vehicle power distribution method provided by an embodiment of the present application;

[0058] Figure 3 is a flowchart of another vehicle power distribution method provided by the present application;

[0059] Figure 4 is a device block diagram of a vehicle power distribution device provided in an embodiment of the present application;

[0060] Figure 5 is a structural diagram of an electronic device provided in an embodiment of the present application;

[0061] Figure 6 It is an overall schematic diagram of vehicle power distribution provided by an embodiment of the present application;

[0062] Figure 7 It is a schematic diagram of constructing a key parameter graph structure provided in an embodiment of the present application;

[0063] Figure 8 It is a schematic diagram of a key neighborhood selection mechanism and a high-order graph generation process provided by an embodiment of the present application;

[0064] Fig. 9 It is a schematic diagram of a time dimension graph convolution structure provided in an embodiment of the present application;

[0065] Fig.10 It is a flow chart of a power allocation scheme based on a genetic algorithm provided in an embodiment of the present application. DETAILED DESCRIPTION

[0066] The exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0067] Figure 1 is a flow chart of the steps of a vehicle power distribution method provided by an embodiment of the present application, such as Figure 1 As shown, the method may include:

[0068] Step 101, constructing a key parameter graph structure according to the correlation between the key characteristic parameters corresponding to the target battery of the target vehicle and the number of cycles, wherein the key parameter graph structure includes an adjacency matrix and node eigenvalues, the adjacency matrix is ​​generated based on the multiplication of an initial characteristic matrix of each of the key characteristic parameters and a transposed matrix corresponding to the initial characteristic matrix, and the node eigenvalues ​​include the key characteristic parameters;

[0069] It should be noted that in the embodiment of the present application, the graph convolutional neural network (GCN) reflects the spatial features in the form of a graph by deeply mining the interdependence between feature nodes in the GCN, which can well mine the hidden information between the features.

[0070] Therefore, in the present application, a key parameter graph structure can be constructed for the corresponding relationship between the key characteristic parameters of the battery and time, that is, the number of cycles.

[0071] Further, in the embodiments of the present application, Figure 3 As shown, the key parameter graph structure is constructed according to the correlation between the key characteristic parameters corresponding to the target battery of the target vehicle and the number of cycles, including:

[0072] Step 1011, generating a constant current charging time variation curve according to the constant current charging time and the number of cycles, generating a constant voltage charging time variation curve according to the constant voltage charging time and the number of cycles, and generating an internal resistance variation curve according to the internal resistance and the number of cycles;

[0073] Among them, the constant current charging time (CCCT), constant voltage charging time (CVCT) and the relationship between internal resistance and cycle number are the key characteristic parameters of battery life, such as Figure 6 As shown, in the embodiment of the present application, three key characteristic parameters of battery life are first selected, namely constant current charging time (CCCT), constant voltage charging time (CVCT) and the relationship between internal resistance and number of cycles.

[0074] Therefore, the constant current charging time variation curve, the constant voltage charging time variation curve, and the internal resistance variation curve can be obtained according to the above relationship.

[0075] Step 1012, dividing the number of cycles according to the constant current charging time variation curve, the constant voltage charging time variation curve, and the internal resistance variation curve according to preset window values ​​and performing feature extraction to obtain an initial feature matrix;

[0076] The three key feature parameters are divided into n+1 segments according to the window size T according to the number of cycles. Each segment contains three key feature parameters corresponding to the number of cycles. Then, the multi-layer perceptron MLP is used to extract features of each key feature parameter in each segment, and the initial feature matrix corresponding to each key feature parameter can be obtained.

[0077] Step 1013, generating an adjacency matrix based on multiplying the initial feature matrix of each of the key feature parameters by the transposed matrix corresponding to the initial feature matrix;

[0078] Specifically, the number of cycles is divided into n+1 segments according to the preset window value T. Taking the nth segment as an example, H in The features extracted from each key feature parameter in the nth segment can be specifically expressed as:

[0079] H in =MLP in (X in )(Formula 1)

[0080] In the above formula 1, i = 1, 2, 3, corresponding to CCCT, CVCT and internal resistance respectively, X in Represents the three key feature parameters in the nth segment, and MLP is a multi-layer perceptron function.

[0081] Then, the feature matrix extracted based on formula 1 is multiplied by its transposed matrix to obtain the adjacency matrix, where the adjacency matrix can be expressed as:

[0082]

[0083] Step 1014: construct a key parameter graph structure according to the initial feature matrix and the adjacency matrix.

[0084] Through the above operation, an adjacency matrix is ​​constructed for each key characteristic parameter in each segment, and the corresponding parameters are used as the eigenvalues ​​of the node. The obtained adjacency matrix and the eigenvalues ​​of the node can be obtained in the form of a graph G in (Xi n, A in), where the key parameter graph structure construction process is as follows Figure 7 shown.

[0085] Step 102, generating a battery life prediction value according to a pre-constructed battery life prediction model and the key parameter graph structure, wherein the pre-constructed battery life prediction model is constructed based on the key parameter graph structure and the target enhanced feature;

[0086] It should be noted that in the embodiment of the present application, after the key parameter graph structure corresponding to the three key parameters is constructed, a spatiotemporal graph convolution battery life prediction model can be constructed based on this, that is, the pre-constructed battery life prediction model in the present application.

[0087] Among them, the battery life prediction model includes spatial feature extraction of key characteristic parameters of the battery in the spatial dimension and enhanced feature recursive extraction in the time dimension.

[0088] Furthermore, after realizing spatial feature extraction and enhanced feature recursive extraction, a battery life prediction model can be constructed based on the final enhanced features, and then the final battery life prediction output can be obtained.

[0089] Step 103: Control the target vehicle to distribute power according to the battery life prediction value.

[0090] It should be noted that, in the embodiment of the present application, in the above step 103, after the battery life prediction value is determined, the power distribution of the target vehicle can be controlled by the battery life prediction value.

[0091] Specifically, the target power distribution parameter can be determined according to the battery life prediction value and the pre-acquired minimum fuel consumption strategy, and then the target vehicle can be controlled to distribute power according to the power distribution parameter.

[0092] First, the battery life prediction value can be combined with the minimum fuel consumption strategy based on genetic algorithm.

[0093] Genetic algorithms can be applied to the optimization of power distribution strategies. By continuously evolving and optimizing power distribution parameters, the best power distribution plan can be found. At the same time, the battery life prediction results are introduced into the genetic algorithm to achieve minimum fuel consumption while taking into account the limitations of battery life.

[0094] Among them, the minimum fuel consumption strategy can be an equivalent consumption minimization strategy (ECMS). ECMS can produce instantaneous optimal control, so the calculation efficiency is very high. ECMS can minimize fuel consumption under certain conditions. Specifically, this application does not make specific limitations on this.

[0095] Furthermore, the determining of the target power distribution parameter according to the battery life prediction value and the pre-acquired minimum fuel consumption strategy includes:

[0096] Determine an equivalent fuel consumption value according to a pre-acquired minimum fuel consumption strategy, and use the equivalent fuel consumption value as a first objective function and the battery life prediction value as a second objective function;

[0097] The target power distribution parameter is determined by a preset genetic algorithm, the first objective function and the second objective function.

[0098] It should be noted that the reference Fig.10,The first step is to determine the optimization target, with the vehicle’s equivalent fuel economy and the battery ,cycle life as the objective function of the system.

[0099] Among them, the first objective function corresponding to the equivalent fuel of the whole vehicle is expressed as Among them, p req -p eng is the difference between the required power and the engine power, that is, the battery power, because the required power is provided by the engine and the battery; s is the equivalent factor, is the instantaneous fuel consumption of the engine power corresponding to the working point, and Qf is the lower calorific value of the engine fuel (J / g).

[0100] The second objective function corresponding to the battery cycle life is expressed as min|y pre -y curr |; where ypred is the predicted battery life; ycurr is the current battery life. The change in the difference can be used to measure the effect of the energy management strategy on the optimization of power distribution.

[0101] If the difference after optimization is small, it means that the optimization strategy successfully makes the predicted battery life close to the current battery life, indicating that the optimization of the energy management strategy plays a positive role in extending the battery life.

[0102] Then a genetic algorithm method integrating elite strategy is used to solve the problem. The elite strategy is to obtain the dominance level Zi by mixing the parent population with its offspring population and performing fast non-dominated sorting. A crowding comparison operator is introduced, and the crowding degree di is used as the comparison criterion between individuals in the population. By comparing Zi and di, excellent individuals are selected to maintain population diversity.

[0103] Among them, the simplified model of NSGA can be expressed as:

[0104] NSGA = {C, J, POP, GEN, M, N, V, A, B, C} (Formula 3)

[0105] Among them, in the above formula 3: C is an individual encoding method, which converts the power thresholds X1, X2, X3, X4, SOC thresholds and other variables to be optimized into string genes in a binary manner, and concatenates them into a chromosome, that is, a vector X in the optimization solution space represents a chromosome encoding: J is an individual fitness evaluation function, with minJ=O(Z) as the objective function: POP is the initial population, and the initial population is generated by a consistent random method: M is the number of objective functions, which is set to 2: N is the number of populations, V is the dimension, which represents the number of decision variables, and a value is randomly selected from the maximum and minimum values ​​as each decision variable; A selects the operator: B is the mutation operator, which uses the basic bit mutation operator: C is the operation stop condition algebra, which is set to stop the operation when the maximum number of iterations is reached.

[0106] In an embodiment of the present application, a key parameter graph structure is constructed according to the correlation between the key characteristic parameters corresponding to the target battery of the target vehicle and the number of cycles, wherein the key parameter graph structure includes an adjacency matrix and node eigenvalues, wherein the adjacency matrix is ​​generated based on the multiplication of the initial characteristic matrix of each of the key characteristic parameters and the transposed matrix corresponding to the initial characteristic matrix, and the node eigenvalues ​​include the key characteristic parameters; a battery life prediction value is generated according to a pre-constructed battery life prediction model and the key parameter graph structure, wherein the pre-constructed battery life prediction model is constructed based on the key parameter graph structure and target enhancement features; a target power distribution parameter is determined according to the battery life prediction value and a pre-acquired minimum fuel consumption strategy; and the target vehicle is controlled to distribute power according to the power distribution parameter. That is, the embodiment of the present application selects key characteristic parameters related to battery life, constructs a key parameter graph structure based on the key characteristic parameters, and then combines the constructed battery life prediction model with the minimum fuel consumption strategy, and determines the optimal target power distribution parameter by continuously evolving and optimizing the power distribution parameter, and then achieves the best balance between fuel consumption and battery life according to the power distribution scheme and the limitation of battery life while considering the battery life.

[0107] Figure 2 is a flow chart of the steps of a vehicle power distribution method provided by an embodiment of the present application, such as Figure 2 As shown, the method may include:

[0108] Step 201, constructing a key parameter graph structure according to the correlation between the key characteristic parameters corresponding to the target battery of the target vehicle and the number of cycles, wherein the key parameter graph structure includes an adjacency matrix and node eigenvalues, the adjacency matrix is ​​generated based on the multiplication of an initial characteristic matrix of each of the key characteristic parameters and a transposed matrix corresponding to the initial characteristic matrix, and the node eigenvalues ​​include the key characteristic parameters;

[0109] It should be noted that, in the embodiment of the present application, the above step 201 refers to the previous discussion and will not be repeated here.

[0110] Step 202, extracting spatial features from the key parameter graph structure to obtain initial spatial features;

[0111] Step 203, sorting the adjacency matrix to obtain a target adjacency matrix;

[0112] Step 204, generating a first key neighborhood graph according to the initial spatial features and the target adjacency matrix;

[0113] Step 205: generating a high-order graph structure according to the first key neighborhood graph, wherein the high-order graph structure represents the target space features corresponding to the key feature parameters;

[0114] Step 206, performing key neighborhood extraction on the high-order graph structure to obtain a second key neighborhood graph, wherein the second key neighborhood graph includes target space features corresponding to key neighborhood graph nodes extracted from the high-order graph structure;

[0115] Step 207, performing enhancement processing on the second spatial feature according to a pre-constructed enhancement feature formula to obtain a target enhancement feature;

[0116] Step 208, generating a battery life prediction model according to the target enhanced features;

[0117] It should be noted that, in the embodiment of the present application, in the above steps 202-208, spatial feature extraction is first performed, and the specific steps are as follows:

[0118] The key parameter graph structure G in (Xi n, Ai n) constructed by the relationship between CCCT, CVCT and internal resistance and cycle number is extracted. First, the graph structure is preliminarily extracted by graph convolution. The extracted features are used to extract the features. Indicates that the graph convolution calculation process is as follows:

[0119]

[0120] In the above formula 4, i and n have the same meaning as above, and W iin is the corresponding weight matrix, is the output of the ith key feature of the nth segment, In order to perform symmetric normalization on the adjacency matrix Ai n, ReLU is used as the activation function.

[0121] Then a key neighborhood selection mechanism is designed to select the most representative key points and their adjacent nodes within each key feature.

[0122] By performing the Max sorting operation on the adjacency matrix Ain, the first 5 maximum values ​​of each row in the adjacency matrix Ain are retained, and the rest are set to 0, and the adjacency matrix Ain is obtained. i ' n , recorded as:

[0123] A i ' n =max k (A in )(Formula 5)

[0124] Among them, in this A i ' n In the matrix, each row represents the connection relationship between a node and other nodes. The sum of each row is calculated to obtain the row index of the maximum value, that is, the most representative key node. At the same time, the top 5 nodes with the maximum value connected to it are extracted to form a key neighborhood graph. Through the above operations, the key neighborhood graph corresponding to the three features will be obtained in each segment, which is recorded as

[0125]

[0126] in, Represents the features corresponding to the key neighborhood graph nodes, The adjacency matrix corresponding to the key neighborhood of the ith feature in the nth segment. By selecting the most representative key points and their adjacent nodes in each key feature, the most important features can be extracted from the preliminary features, which helps to reduce the dimension of the data and remove redundant information, thereby improving the effect of subsequent analysis and prediction.

[0127] Then the three key neighborhood graphs in each segment obtained above are used to generate the high-order graph structure of each segment, and the interdependence between the key battery parameters in the target segment is deeply mined to reflect the spatial characteristics. The key neighborhood selection mechanism and the high-order graph generation process are as follows: Figure 8 The specific steps are as follows:

[0128] Suppose that the three key neighborhood graphs of a certain section are G1=(X1, A1), and the most representative key node is denoted by id1; G2=(X2, A2), and the most representative key node is denoted by id2; G3=(X3, A3), and the most representative key node is denoted by id3.

[0129] Connect id1, id2 and id3 in pairs. In this way, the key nodes in the three key neighborhood graphs are connected. The key nodes are attached to the key points corresponding to their respective neighborhoods. In this way, the three key neighborhood graphs in a certain section are combined to generate a high-order graph structure, which is represented by Gn=(Xn,An).

[0130] The high-order graph structure can establish more complex relationships by connecting different key nodes. This helps to enhance the ability of feature representation, enabling the model to better capture the complex associations and nonlinear relationships between key battery parameters.

[0131] The generated high-order graph is then subjected to three layers of Chebyshev graph convolution. Chebyshev graph convolution is a graph convolution operation based on the adjacency matrix, which updates the representation of the node through multiple layers of adjacency matrix multiplication and feature propagation.

[0132] By applying 3 layers of Chebyshev graph convolution, multiple feature propagations and updates can be performed in the high-order graph structure, thereby capturing longer-range dependencies between nodes and generating more complex and richer feature representations.

[0133] The specific calculation process is as follows:

[0134] H n = Cheb(σ(Cheb(σ(Cheb(X n ,W (0) )),W (1) )),W (2) ))(Formula 7)

[0135] In the above formula 7, Hn represents the aggregated features generated by the n-th layer high-order graph structure, σ is the function ReLu, and W(0), W(1) and W(2) are the weight matrices corresponding to the Chebyshev convolutional layer.

[0136] Then, for each window, the key information of each period is mined through the key neighborhood selection mechanism as the enhanced features of the next moment, effectively capturing the changing pattern of battery key information in the time dimension. The time dimension graph convolution structure is as follows: Fig. 9 As shown, the specific operations are as follows:

[0137] First, based on the above key neighborhood selection mechanism, the key neighborhood extraction is performed on the high-order graph structure Gn=(Xn,An). In this An matrix, each row is summed to obtain the row index of the maximum value, that is, the most representative key node in the nth segment of the high-order graph is obtained. At the same time, the top 5 nodes of the maximum value connected to it are extracted to form a key neighborhood graph of the nth segment of the high-order graph. Through the above operation, the corresponding key neighborhood graph will be obtained in each segment, denoted as

[0138]

[0139] Among them, in the above formula 8, Represents the features corresponding to the key neighborhood graph nodes extracted from the nth segment of the high-order graph, Represents the adjacency matrix corresponding to the key neighborhood of the nth segment of the high-order graph.

[0140] Then, the features corresponding to the key neighborhood graph obtained in the nth segment and the key neighborhood graph obtained in the n+1 segment are combined, and then the feature enhancement mechanism is used to mine the relevant information between the nth segment and the next segment n+1 to generate enhanced features to improve the ability to extract key features in the n+1 segment. The enhanced feature expression is as follows:

[0141]

[0142] Among them, in the above formula 9, f is the si gmo id function and is the re lu activation function.

[0143] In summary, a good battery life prediction model is constructed.

[0144] Step 209, generating a battery life prediction value according to a pre-constructed battery life prediction model and the key parameter graph structure, wherein the pre-constructed battery life prediction model is constructed based on the key parameter graph structure and the target enhanced feature;

[0145] Finally, the enhanced features Through a fully connected layer, the final battery life prediction output yprediction is obtained.

[0146] Step 210: Control the target vehicle to distribute power according to the battery life prediction value.

[0147] It should be noted that the above step 210 is referred to the previous discussion and will not be repeated here.

[0148] In an embodiment of the present application, a key parameter graph structure is constructed according to the correlation between the key characteristic parameters corresponding to the target battery of the target vehicle and the number of cycles, wherein the key parameter graph structure includes an adjacency matrix and node eigenvalues, wherein the adjacency matrix is ​​generated based on the multiplication of the initial characteristic matrix of each of the key characteristic parameters and the transposed matrix corresponding to the initial characteristic matrix, and the node eigenvalues ​​include the key characteristic parameters; a battery life prediction value is generated according to a pre-constructed battery life prediction model and the key parameter graph structure, wherein the pre-constructed battery life prediction model is constructed based on the key parameter graph structure and target enhancement features; a target power distribution parameter is determined according to the battery life prediction value and a pre-acquired minimum fuel consumption strategy; and the target vehicle is controlled to distribute power according to the power distribution parameter. That is, the embodiment of the present application selects key characteristic parameters related to battery life, constructs a key parameter graph structure based on the key characteristic parameters, and then combines the constructed battery life prediction model with the minimum fuel consumption strategy, and determines the optimal target power distribution parameter by continuously evolving and optimizing the power distribution parameter, and then achieves the best balance between fuel consumption and battery life according to the power distribution scheme and the limitation of battery life while considering the battery life.

[0149] In addition, the spatiotemporal graph convolution battery life prediction model mines the key characteristic parameters of the battery from the two dimensions of time and space, and can efficiently extract the relationship between the key characteristic parameters, help discover potential factors affecting battery life, and use them to establish a prediction model.

[0150] Among them, the above model establishment process involves a key neighborhood selection mechanism, which selects the most representative key points and their adjacent nodes in each key feature to generate a high-order graph structure, deeply mine the interdependence between key battery parameters to reflect spatial characteristics; and, for each window, the key information of each time period is mined through the key neighborhood selection mechanism as the enhanced feature of the next moment, effectively capturing the changing pattern of battery key information in the time dimension, helping the prediction model to better understand the changes in battery life.

[0151] Corresponding to the method provided in the above-mentioned vehicle power distribution method embodiment of the present application, see Figure 4 The present application also provides a logic block diagram of a vehicle power distribution device. In this embodiment, the device 400 includes:

[0152] A key parameter graph structure construction module 401 is used to construct a key parameter graph structure according to the correlation between the key characteristic parameters corresponding to the target battery of the target vehicle and the number of cycles, wherein the key parameter graph structure includes an adjacency matrix and node eigenvalues, the adjacency matrix is ​​generated based on the multiplication of an initial characteristic matrix of each of the key characteristic parameters and a transposed matrix corresponding to the initial characteristic matrix, and the node eigenvalue includes the key characteristic parameter;

[0153] A generating module 402, configured to generate a battery life prediction value according to a pre-constructed battery life prediction model and the key parameter graph structure, wherein the pre-constructed battery life prediction model is constructed based on the key parameter graph structure and the target enhancement feature;

[0154] The distribution module 403 is used to control the target vehicle to distribute power according to the battery life prediction value.

[0155] Optionally, the device further comprises:

[0156] The first module is used to extract spatial features from the key parameter graph structure to obtain initial spatial features;

[0157] The second module is used to sort the adjacency matrix to obtain a target adjacency matrix;

[0158] A third module is used to generate a first key neighborhood graph according to the initial spatial features and the target adjacency matrix;

[0159] A fourth module is used to generate a high-order graph structure according to the first key neighborhood graph, wherein the high-order graph structure represents the target space features corresponding to the key feature parameters;

[0160] A fifth module is used to extract key neighborhoods for the high-order graph structure to obtain a second key neighborhood graph, wherein the second key neighborhood graph includes target space features corresponding to key neighborhood graph nodes extracted from the high-order graph structure;

[0161] A sixth module is used to enhance the second spatial feature according to a pre-constructed enhancement feature formula to obtain a target enhancement feature;

[0162] The seventh module is used to generate a battery life prediction model according to the target enhanced features.

[0163] Optionally, the generating module includes:

[0164] The generation submodule is used to output the battery life prediction value through a fully connected layer through the target enhanced features.

[0165] Optionally, the key parameter graph structure building module includes:

[0166] A first construction submodule is used to generate a constant current charging time variation curve according to the constant current charging time and the number of cycles, to generate a constant voltage charging time variation curve according to the constant voltage charging time and the number of cycles, and to generate an internal resistance variation curve according to the internal resistance and the number of cycles;

[0167] A second construction submodule is used to divide the number of cycles according to the constant current charging time variation curve, the constant voltage charging time variation curve and the internal resistance variation curve according to preset window values ​​and perform feature extraction to obtain an initial feature matrix;

[0168] A third construction submodule is used to generate an adjacency matrix based on multiplying an initial feature matrix of each of the key feature parameters by a transposed matrix corresponding to the initial feature matrix;

[0169] The fourth construction submodule is used to construct a key parameter graph structure according to the initial feature matrix and the adjacency matrix.

[0170] Optionally, the determining module includes:

[0171] A first determination submodule, configured to determine an equivalent fuel consumption value according to a pre-acquired minimum fuel consumption strategy, and use the equivalent fuel consumption value as a first objective function and use the battery life prediction value as a second objective function;

[0172] The second determination submodule is used to determine the target power distribution parameter by using a preset genetic algorithm, the first objective function and the second objective function.

[0173] In summary, a vehicle power distribution device provided by an embodiment of the present application constructs a key parameter graph structure according to the correlation between the key characteristic parameters corresponding to the target battery of the target vehicle and the number of cycles, wherein the key parameter graph structure includes an adjacency matrix and a node eigenvalue, wherein the adjacency matrix is ​​generated by multiplying the initial characteristic matrix of each of the key characteristic parameters by the transposed matrix corresponding to the initial characteristic matrix, and the node eigenvalue includes the key characteristic parameter; generates a battery life prediction value according to a pre-constructed battery life prediction model and the key parameter graph structure, wherein the pre-constructed battery life prediction model is constructed based on the key parameter graph structure and the target enhanced feature; determines a target power distribution parameter according to the battery life prediction value and a pre-acquired minimum fuel consumption strategy; and controls the target vehicle to distribute power according to the power distribution parameter. That is, the embodiment of the present application selects key characteristic parameters related to battery life, constructs a key parameter graph structure based on the key characteristic parameters, and then combines the constructed battery life prediction model with the minimum fuel consumption strategy, determines the optimal target power distribution parameter by continuously evolving and optimizing the power distribution parameter, and then achieves the best balance between fuel consumption and battery life according to the power distribution scheme and the limitation of battery life while considering the battery life.

[0174] Figure 5 This is a structural diagram of an electronic device M00 provided in an embodiment of the present application. In the figure, the electronic device M00 includes a processor M01 and a memory M02. The memory M02 stores programs or instructions that can be run on the processor M01. When the program or instructions are executed by the processor M01, the various steps of the above-mentioned vehicle power distribution method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, they will not be repeated here.

[0175] In an embodiment of the present application, the memory M02 can be used to store software programs and various data. The memory M02 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instruction required for at least one function (such as a sound playback function, an image playback function, etc.), etc. In addition, the memory M02 may include a volatile memory or a non-volatile memory, or the memory x09 may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. Volatile memory can be random access memory (Random Access Memory, RAM), static random access memory (Static RAM, SRAM), dynamic random access memory (Dynamic RAM, DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDRSDRAM), enhanced synchronous dynamic random access memory (EnhancedSDRAM, ESDRAM), synchronous link dynamic random access memory (Synch link DRAM, SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DRRAM). The memory M02 in the embodiment of the present application includes but is not limited to these and any other suitable types of memory.

[0176] The processor M01 may include one or more processing units; optionally, the processor M01 integrates an application processor and a modem processor, wherein the application processor mainly processes operations related to an operating system, a user interface, and application programs, and the modem processor mainly processes wireless communication signals, such as a baseband processor. It is understandable that the modem processor may not be integrated into the processor M01.

[0177] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various processes of the above-mentioned vehicle power distribution method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0178] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.

[0179] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned vehicle power distribution method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0180] It should be understood that the chip involved in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0181] An embodiment of the present application provides a computer program product, which is stored in a storage medium. The program product is executed by at least one processor to implement the various processes of the above-mentioned vehicle power distribution method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0182] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0183] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the relevant technology, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0184] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A vehicle power distribution method, characterized in that: The method comprises: Constructing a key parameter graph structure according to the correlation between the key characteristic parameters corresponding to the target battery of the target vehicle and the number of cycles, wherein the key parameter graph structure includes an adjacency matrix and node eigenvalues, the adjacency matrix is ​​generated based on the multiplication of an initial characteristic matrix of each of the key characteristic parameters and a transposed matrix corresponding to the initial characteristic matrix, and the node eigenvalues ​​include the key characteristic parameters; generating a battery life prediction value according to a pre-constructed battery life prediction model and the key parameter graph structure, wherein the pre-constructed battery life prediction model is constructed based on the key parameter graph structure and the target enhancement feature; The target vehicle is controlled to distribute power according to the battery life prediction value.

2. The method according to claim 1, characterized in that Before the step of generating a battery life prediction value according to the pre-built battery life prediction model and the key parameter graph structure, the method includes: Extracting spatial features from the key parameter graph structure to obtain initial spatial features; Sorting the adjacency matrix to obtain a target adjacency matrix; Generating a first key neighborhood graph according to the initial spatial features and the target adjacency matrix; Generating a high-order graph structure according to the first key neighborhood graph, wherein the high-order graph structure represents the target space features corresponding to the key feature parameters; Performing key neighborhood extraction on the high-order graph structure to obtain a second key neighborhood graph, wherein the second key neighborhood graph includes target space features corresponding to key neighborhood graph nodes extracted from the high-order graph structure; Performing enhancement processing on the second spatial feature according to a pre-constructed enhancement feature formula to obtain a target enhanced feature; A battery life prediction model is generated according to the target enhanced features.

3. The method according to claim 2, characterized in that The generating of the battery life prediction value according to the pre-built battery life prediction model and the key parameter graph structure comprises: The target enhanced features are passed through a fully connected layer to output a battery life prediction value.

4. The method according to claim 1, characterized in that: The key characteristic parameters include at least one of constant current charging time, constant voltage charging time, and internal resistance.

5. The method according to claim 5, characterized in that: The key parameter graph structure is constructed according to the correlation between the key characteristic parameters corresponding to the target battery of the target vehicle and the number of cycles, including: Generating a constant current charging time variation curve according to the constant current charging time and the number of cycles, generating a constant voltage charging time variation curve according to the constant voltage charging time and the number of cycles, and generating an internal resistance variation curve according to the internal resistance and the number of cycles; According to the constant current charging time variation curve, the constant voltage charging time variation curve and the internal resistance variation curve, the number of cycles is divided according to preset window values ​​and features are extracted to obtain an initial feature matrix; Generate an adjacency matrix based on the multiplication of an initial feature matrix of each of the key feature parameters and a transposed matrix corresponding to the initial feature matrix; A key parameter graph structure is constructed according to the initial feature matrix and the adjacency matrix.

6. The method according to claim 1, characterized in that The step of determining the target power distribution parameter according to the battery life prediction value and the pre-acquired minimum fuel consumption strategy includes: Determine an equivalent fuel consumption value according to a pre-acquired minimum fuel consumption strategy, and use the equivalent fuel consumption value as a first objective function and the battery life prediction value as a second objective function; The target power distribution parameter is determined by a preset genetic algorithm, the first objective function and the second objective function.

7. A vehicle power distribution device, characterized in that: The device comprises: A key parameter graph structure construction module, used to construct a key parameter graph structure according to the correlation between the key characteristic parameters corresponding to the target battery of the target vehicle and the number of cycles, wherein the key parameter graph structure includes an adjacency matrix and node eigenvalues, the adjacency matrix is ​​generated based on the multiplication of an initial characteristic matrix of each of the key characteristic parameters and a transposed matrix corresponding to the initial characteristic matrix, and the node eigenvalues ​​include the key characteristic parameters; A generating module, configured to generate a battery life prediction value according to a pre-constructed battery life prediction model and the key parameter graph structure, wherein the pre-constructed battery life prediction model is constructed based on the key parameter graph structure and the target enhancement feature; A distribution module is used to control the target vehicle to distribute power according to the battery life prediction value.

8. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; Wherein, the processor is configured to execute the instructions to implement the vehicle power distribution method as described in any one of claims 1 to 6.

9. A storage medium, characterized in that: When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the vehicle power distribution method as claimed in any one of claims 1 to 6.

10. A vehicle, characterized in that: The vehicle includes the vehicle power distribution method as claimed in claim 7.