Range extender temperature prediction model training method, range extender control method and controller

Abstract

The invention relates to a range extender temperature prediction model training method, a range extender control method and a controller. The training method comprises the following steps: acquiring sample range extender temperature data corresponding to different time windows under multiple working conditions, and constructing a sample range extender temperature feature map; inputting the sample range extender temperature feature graph into a range extender temperature prediction model to be trained, extracting graph convolution features of different feature dimensions, and obtaining a feedback coefficient of each feature dimension; according to each feedback coefficient, obtaining an active selection factor of each feature dimension, and obtaining an optimal feedback coefficient; utilizing the active selection factor of each feature dimension to fuse the graph convolution features of each feature dimension, and performing feature enhancement based on an optimal feedback coefficient to obtain time enhancement features; and obtaining predicted range extender temperature data based on the time enhancement features to obtain a trained range extender temperature prediction model. By adopting the method, the accuracy of temperature prediction of the range extender can be improved.

Description

Technical Field

[0001] The present invention relates to the field of vehicle control technology, and in particular to a range extender temperature prediction model training method, a range extender control method and a controller. Background Art

[0002] With the development of vehicle control technology, a technology has emerged that uses a range extender to provide power to the vehicle when the battery is low, thereby extending the driving range. When the battery of an electric vehicle is low, the range extender automatically starts to power the battery, allowing the vehicle to continue driving. In addition, the temperature control of the range extender is one of the key factors to ensure its normal operation and extend its life. Therefore, accurate prediction and real-time monitoring of the temperature changes of the range extender are crucial to ensure its stability and performance.

[0003] In related technologies, temperature prediction of range extenders is often limited by model complexity and real-time performance, making it difficult to meet the demands for high-precision and real-time temperature prediction of range extenders. However, with the rapid development of artificial intelligence technology, through learning and pattern recognition of large amounts of data, it is possible to explore the complex relationship between range extender temperature and various factors, achieve efficient and accurate temperature prediction, and provide strong support for performance optimization and safe operation of electric vehicles.

[0004] However, the current range extender temperature prediction achieved through artificial intelligence technology usually requires a large number of high-quality samples for training models. In addition, it also faces the problem of balancing model complexity and real-time performance, as well as challenges in model generalization and stability. Therefore, the currently trained range extender temperature prediction model has low accuracy in predicting range extender temperature. Summary of the invention

[0005] Based on this, it is necessary to provide a range extender temperature prediction model training method, range extender control method, device, vehicle controller, storage medium and computer program product that can improve the accuracy of range extender temperature prediction in response to the above technical problems.

[0006] In a first aspect, the present invention provides a range extender temperature prediction model training method, comprising:

[0007] Acquire sample range extender temperature data corresponding to different time windows under multiple operating conditions, and construct a sample range extender temperature characteristic diagram based on the sample range extender temperature data of each of the time windows;

[0008] Inputting the sample range extender temperature characteristic graph into the range extender temperature prediction model to be trained, extracting graph convolution features of the sample range extender temperature characteristic graph corresponding to different feature dimensions through the range extender temperature prediction model, and obtaining feedback coefficients of each feature dimension based on each graph convolution feature;

[0009] According to each of the feedback coefficients, an active selection factor of each of the feature dimensions is obtained, and an optimal feedback coefficient is obtained from each of the feedback coefficients;

[0010] Using the active selection factor of each feature dimension, the graph convolution features of each feature dimension are fused to obtain a fused feature, and the fused feature is enhanced based on the optimal feedback coefficient to obtain a time-enhanced feature;

[0011] Predicted range extender temperature data is acquired based on the time enhancement feature, and the range extender temperature prediction model is trained using the predicted range extender temperature data to obtain a trained range extender temperature prediction model.

[0012] In one embodiment, the feedback coefficient of each feature dimension is composed of multiple sub-feedback coefficients, and each sub-feedback coefficient corresponds to a different time window; the active selection factor of each feature dimension is obtained according to each feedback coefficient, including: obtaining the current feature dimension, and the current sub-feedback coefficient of the current feature dimension corresponding to each time window; the current feature dimension is any one of the feature dimensions; according to a preset feedback coefficient threshold and each current sub-feedback coefficient, the target time window corresponding to the current feature dimension and the number of target time windows are obtained from each time window; the current sub-feedback coefficient corresponding to the target time window is less than or equal to the feedback coefficient threshold; and the ratio of the number of target time windows to the total number of windows of each time window is used as the active selection factor of the current feature dimension.

[0013] In one of the embodiments, the feedback coefficient of each feature dimension is obtained based on each graph convolution feature, including: inputting the graph convolution feature of the current feature dimension into the fully connected layer of the range extender temperature prediction model, and obtaining the initial predicted range extender temperature data of the current feature dimension corresponding to each time window through the fully connected layer; obtaining the difference between the initial predicted range extender temperature data of each time window and the sample range extender temperature data of each time window, and normalizing the difference of each time window to obtain each current sub-feedback coefficient.

[0014] In one embodiment, the active selection factor of each feature dimension is used to fuse the graph convolution features of each feature dimension to obtain the fused feature, including: obtaining the fusion weight of each feature dimension according to the active selection factor of each feature dimension; wherein the size of the fusion weight is positively correlated with the size of the active selection factor; and using the fusion weight of each feature dimension to weightedly process the graph convolution features of each feature dimension to obtain the fused feature.

[0015] In one embodiment, obtaining the optimal feedback coefficient from each of the feedback coefficients includes: obtaining a target feature dimension, and taking the feedback coefficient corresponding to the target feature dimension as the optimal feedback coefficient; the active selection factor corresponding to the target feature dimension is the maximum value of each of the active selection factors; and enhancing the fused feature based on the optimal feedback coefficient to obtain a time-enhanced feature includes: normalizing the optimal feedback coefficient, and using the normalized optimal feedback coefficient to enhance the fused feature to obtain the time-enhanced feature.

[0016] In one embodiment, the fusion feature is composed of multiple sub-fusion features, each of which corresponds to a different time window; the time enhancement feature is composed of multiple sub-time enhancement features, each of which corresponds to a different time window; the acquisition of predicted range extender temperature data based on the time enhancement feature includes: acquiring the sub-time enhancement feature corresponding to the current time window, and the sub-time enhancement feature and sub-fusion feature of the previous time window of the current time window; the current time window is any one of the time windows; using the sub-time enhancement feature corresponding to the current time window, and the sub-time enhancement feature and sub-fusion feature of the previous time window, obtain the timing enhancement feature corresponding to the current time window; input the timing enhancement feature into the fully connected layer of the range extender temperature prediction model, and obtain the predicted range extender temperature data of the next time window of the current time window through the fully connected layer.

[0017] In one of the embodiments, the sample range extender temperature characteristic graph is constructed based on the sample range extender temperature data of each time window, including: taking each time window as a graph node of the sample range extender temperature characteristic graph, and taking the sample range extender temperature data of each time window as a node feature of each graph node; using each node feature to obtain the similarity between each graph node, and according to the similarity, obtaining an adjacency matrix for characterizing the connection relationship between each graph node; and constructing the sample range extender temperature characteristic graph according to the adjacency matrix and each node feature.

[0018] In a second aspect, the present invention further provides a range extender control method, comprising:

[0019] Acquire historical range extender temperature data corresponding to different time windows of the target range extender, and construct a historical range extender temperature characteristic diagram based on the historical range extender temperature data of each of the time windows;

[0020] Inputting the historical range extender temperature characteristic diagram into a trained range extender temperature prediction model, and outputting predicted range extender temperature data of the target range extender through the range extender temperature prediction model; wherein the range extender temperature prediction model is trained by the range extender temperature prediction model training method as described in any one of the embodiments of the first aspect;

[0021] The target range extender is temperature controlled using the predicted range extender temperature data.

[0022] In a third aspect, the present invention further provides a range extender temperature prediction model training device, comprising:

[0023] A sample characteristic diagram construction module, used to obtain sample range extender temperature data corresponding to different time windows under multiple working conditions, and to construct a sample range extender temperature characteristic diagram based on the sample range extender temperature data of each of the time windows;

[0024] A feedback coefficient acquisition module, used for inputting the sample range extender temperature characteristic graph into the range extender temperature prediction model to be trained, extracting graph convolution features of the sample range extender temperature characteristic graph corresponding to different feature dimensions through the range extender temperature prediction model, and acquiring feedback coefficients of each feature dimension based on each graph convolution feature;

[0025] A selection factor acquisition module, used to acquire the active selection factor of each of the feature dimensions according to each of the feedback coefficients, and to acquire the optimal feedback coefficient from each of the feedback coefficients;

[0026] An enhanced feature acquisition module, used to fuse the graph convolution features of each feature dimension using the active selection factor of each feature dimension to obtain a fused feature, and to enhance the fused feature based on the optimal feedback coefficient to obtain a time-enhanced feature;

[0027] A prediction model training module is used to obtain predicted range extender temperature data based on the time enhancement feature, and use the predicted range extender temperature data to train the range extender temperature prediction model to obtain a trained range extender temperature prediction model.

[0028] In a fourth aspect, the present invention further provides a range extender control device, comprising:

[0029] A historical characteristic map construction module, used to obtain historical range extender temperature data corresponding to different time windows of the target range extender, and to construct a historical range extender temperature characteristic map based on the historical range extender temperature data of each of the time windows;

[0030] a predicted temperature output module, configured to input the historical range extender temperature characteristic diagram into a trained range extender temperature prediction model, and output predicted range extender temperature data of the target range extender through the range extender temperature prediction model; wherein the range extender temperature prediction model is trained by the range extender temperature prediction model training method as described in any one of the embodiments of the first aspect;

[0031] A range extender temperature control module is used to use the predicted range extender temperature data to perform temperature control on the target range extender.

[0032] In a fifth aspect, the present invention further provides a vehicle controller, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method described in any one of the first aspect or the second aspect when executing the computer program.

[0033] In a sixth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method described in any one of the first aspect or the second aspect are implemented.

[0034] In a seventh aspect, the present invention further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the method described in any one of the first aspect or the second aspect.

[0035] The range extender temperature prediction model training method, range extender temperature prediction method, device, vehicle controller, storage medium and computer program product obtain sample range extender temperature data corresponding to different time windows under multiple working conditions, and construct a sample range extender temperature characteristic graph based on the sample range extender temperature data of each time window; input the sample range extender temperature characteristic graph into the range extender temperature prediction model to be trained, extract the graph convolution features of the sample range extender temperature characteristic graph corresponding to different feature dimensions through the range extender temperature prediction model, and obtain the feedback coefficient of each feature dimension based on each graph convolution feature; according to each feedback coefficient, obtain the active selection factor of each feature dimension, and obtain the optimal feedback coefficient from each feedback coefficient; use the active selection factor of each feature dimension to fuse the graph convolution features of each feature dimension to obtain a fused feature, and enhance the fused feature based on the optimal feedback coefficient to obtain a time enhanced feature; obtain the predicted range extender temperature data based on the time enhanced feature, and use the predicted range extender temperature data to train the range extender temperature prediction model to obtain a trained range extender temperature prediction model. The present invention collects sample range extender temperature data of different time windows under multiple working conditions when training the range extender temperature prediction model, thereby constructing a feature graph and inputting the feature graph into the range extender temperature prediction model to be trained, thereby obtaining graph convolution features of different feature dimensions, and obtaining feedback coefficients of each feature dimension based on the graph convolution features, and then using the feedback coefficients to obtain active selection factors of each feature dimension, thereby using the active selection factors to achieve feature fusion, and using the optimal feedback coefficients to perform feature enhancement, and then achieving range extender temperature prediction after obtaining time enhancement features, thereby completing model training. This method realizes multi-dimensional feature output and fusion, can more comprehensively learn the feature representation of data, and can obtain active selection factors based on the feedback coefficients of each feature dimension, and perform feature fusion based on the active selection factors, and can also use the optimal feedback coefficients to perform feature enhancement. The trained range extender temperature prediction model can accurately predict the range extender temperature, thereby improving the accuracy of range extender temperature prediction, and thereby improving the stability of range extender control. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 A schematic diagram of a flow chart of a range extender temperature prediction model training method in one embodiment;

[0038] Figure 2A schematic diagram of a process for obtaining an active selection factor in one embodiment;

[0039] Figure 3 A schematic diagram of a process for obtaining a feedback coefficient in one embodiment;

[0040] Figure 4 A schematic diagram of a process for obtaining predicted range extender temperature data in one embodiment;

[0041] Figure 5 A schematic diagram of a process for constructing a temperature characteristic diagram of a sample range extender in one embodiment;

[0042] Figure 6 1 is a flow chart of a range extender control method in one embodiment;

[0043] Figure 7 A schematic diagram of the structure of an active sample selectively enhanced temperature prediction model in one embodiment;

[0044] Figure 8 It is a structural block diagram of a range extender temperature prediction model training device in one embodiment;

[0045] Fig. 9 is a structural block diagram of a range extender control device in one embodiment;

[0046] Fig.10 FIG. 4 is a diagram showing the internal structure of a vehicle controller in one embodiment. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0048] In one embodiment, Figure 1 As shown, a range extender temperature prediction model training method is provided. This embodiment uses the method applied to a vehicle controller as an example. In this embodiment, the method includes the following steps:

[0049] Step S101, acquiring sample range extender temperature data corresponding to different time windows under multiple working conditions, and constructing a sample range extender temperature characteristic diagram based on the sample range extender temperature data of each time window.

[0050] The sample range extender temperature data refers to the range extender temperature data used as training samples for training the range extender temperature prediction model, and the sample range extender temperature data may be composed of historical temperature data of the range extender. The sample range extender temperature characteristic graph refers to a graph structure used to characterize the sample range extender temperature characteristics.

[0051] Specifically, when training the range extender temperature prediction model, the vehicle controller can first collect the historical temperature data of the range extender under multiple working conditions, and then divide the historical temperature data of the range extender according to a specific time window, for example, according to the time window T, so as to obtain the sample range extender temperature data corresponding to different time windows, which can be expressed as , where n represents the nth time window. Afterwards, the vehicle controller can further use the sample range extender temperature data corresponding to each time window to construct a graph structure for characterizing the sample range extender temperature characteristics, that is, to construct a sample range extender temperature characteristic graph.

[0052] Step S102: input the sample range extender temperature characteristic graph into the range extender temperature prediction model to be trained, extract the graph convolution features of the sample range extender temperature characteristic graph corresponding to different feature dimensions through the range extender temperature prediction model, and obtain the feedback coefficient of each feature dimension based on each graph convolution feature.

[0053] The range extender temperature prediction model to be trained refers to the range extender temperature prediction model that needs to be trained. The model is mainly used to realize the temperature prediction of the automobile range extender, and the graph convolution feature refers to the feature output by the graph convolution operation of the sample range extender temperature feature graph. The feature can characterize the spatial feature information hidden between time windows, and in this embodiment, the graph convolution feature can correspond to different feature dimensions, for example, it can correspond to 3 feature output dimensions, which are 200, 400 and 600. The feedback coefficient can be used to characterize the prediction accuracy coefficient of each feature dimension. The feedback coefficient can be obtained based on the graph convolution feature of each feature dimension, that is, the graph convolution feature of feature dimension 1 can obtain the feedback coefficient of feature dimension 1, and similarly, the graph convolution feature of feature dimension 2 can obtain the feedback coefficient of feature dimension 2, and so on.

[0054] Specifically, after obtaining the sample range extender temperature characteristic graph, the vehicle controller can input the sample range extender temperature characteristic graph into the range extender temperature prediction model that needs to be trained. The model can extract the graph convolution features of the sample range extender temperature characteristic graph corresponding to different feature dimensions, and then further use the graph convolution features of each feature dimension to obtain the feedback coefficient of each feature dimension.

[0055] Step S103, obtaining active selection factors of each feature dimension according to each feedback coefficient, and obtaining the optimal feedback coefficient from each feedback coefficient;

[0056] In step S104, the graph convolution features of each feature dimension are fused using the active selection factor of each feature dimension to obtain a fused feature, and the fused feature is enhanced based on the optimal feedback coefficient to obtain a time-enhanced feature.

[0057] The active selection factor may refer to the fusion factor used to fuse graph convolution features of different feature dimensions, while the optimal feedback coefficient is the enhancement coefficient used to enhance the fused features, i.e., the fused features, and the time enhancement feature refers to the feature obtained after enhancing the fused features using the optimal feedback coefficient.

[0058] Specifically, after obtaining the feedback coefficient of each feature dimension, the vehicle controller can first obtain the active selection factor of each feature dimension, and determine the optimal feedback coefficient from each feedback coefficient. After that, the active selection factor of each feature dimension can be used to fuse the graph convolution features of each feature dimension. After obtaining the fused feature, the optimal feedback coefficient is used to enhance the fused feature to obtain the time enhanced feature.

[0059] Step S105 , obtaining predicted range extender temperature data based on the time enhancement feature, and using the predicted range extender temperature data to train a range extender temperature prediction model to obtain a trained range extender temperature prediction model.

[0060] The predicted range extender temperature data refers to the predicted data output by the range extender temperature prediction model based on the time enhancement feature. Specifically, after obtaining the time enhancement feature, the range extender temperature prediction model can further output the predicted range extender temperature data through the time enhancement feature. Thereafter, the predicted range extender temperature data can be used to train the range extender temperature prediction model to complete the training of the range extender temperature prediction model, that is, to obtain the trained range extender temperature prediction model.

[0061] In the above-mentioned range extender temperature prediction model training method, by obtaining sample range extender temperature data corresponding to different time windows under multiple working conditions, and based on the sample range extender temperature data of each time window, a sample range extender temperature characteristic graph is constructed; the sample range extender temperature characteristic graph is input into the range extender temperature prediction model to be trained, and the graph convolution features of the sample range extender temperature characteristic graph corresponding to different feature dimensions are extracted through the range extender temperature prediction model, and the feedback coefficient of each feature dimension is obtained based on each graph convolution feature; according to each feedback coefficient, the active selection factor of each feature dimension is obtained, and the optimal feedback coefficient is obtained from each feedback coefficient; using the active selection factor of each feature dimension, the graph convolution features of each feature dimension are fused to obtain a fused feature, and the fused feature is feature enhanced based on the optimal feedback coefficient to obtain a time enhanced feature; based on the time enhanced feature, the predicted range extender temperature data is obtained, and the range extender temperature prediction model is trained using the predicted range extender temperature data to obtain a trained range extender temperature prediction model. The present invention collects sample range extender temperature data of different time windows under multiple working conditions when training a range extender temperature prediction model, thereby constructing a feature graph and inputting the feature graph into the range extender temperature prediction model to be trained, thereby obtaining graph convolution features of different feature dimensions, and obtaining feedback coefficients of each feature dimension based on the graph convolution features, and then using the feedback coefficients to obtain active selection factors of each feature dimension, thereby using the active selection factors to achieve feature fusion, and using the optimal feedback coefficients to perform feature enhancement, and then achieving range extender temperature prediction after obtaining time enhancement features, thereby completing model training. This method realizes multi-dimensional feature output and fusion, can more comprehensively learn the feature representation of data, and can obtain active selection factors based on the feedback coefficients of each feature dimension, and perform feature fusion based on the active selection factors, and can also use the optimal feedback coefficients to perform feature enhancement. The trained range extender temperature prediction model can accurately predict the range extender temperature, thereby improving the accuracy of range extender temperature prediction, and thereby improving the stability of range extender control.

[0062] In one embodiment, the feedback coefficient of each feature dimension is composed of a plurality of sub-feedback coefficients, each of which corresponds to a different time window; Figure 2 As shown, step S103 may further include:

[0063] Step S201, obtaining the current feature dimension and the current sub-feedback coefficient of each time window corresponding to the current feature dimension; the current feature dimension is any one of the feature dimensions.

[0064] The current feature dimension may refer to any one of the multiple feature dimensions, and the current sub-feedback coefficient refers to the sub-feedback coefficient of the current feature dimension corresponding to each time window. In this embodiment, the feedback coefficient may be composed of multiple sub-feedback coefficients, and each sub-feedback coefficient corresponds to a different time window. Specifically, the vehicle controller may first use any one of the multiple feature dimensions as the current feature dimension, and use the sub-feedback coefficients corresponding to each time window in the feedback coefficient corresponding to the current feature dimension as the current sub-feedback coefficient of each time window.

[0065] Step S202, according to the preset feedback coefficient threshold and each current sub-feedback coefficient, obtain the target time window corresponding to the current feature dimension and the number of target time windows from each time window; the current sub-feedback coefficient corresponding to the target time window is less than or equal to the feedback coefficient threshold.

[0066] The feedback coefficient threshold is a pre-set coefficient threshold, for example, the coefficient threshold may be 0.3, and the target time window corresponding to the current feature dimension may refer to a time window in which the current sub-feedback coefficient is less than or equal to the feedback coefficient threshold. In this embodiment, after obtaining the current sub-feedback coefficient of each time window, each current sub-feedback coefficient may be compared with the feedback coefficient threshold, so that the time window in which the current sub-feedback coefficient is less than or equal to the feedback coefficient threshold is used as the target time window corresponding to the current feature dimension.

[0067] For example, the current sub-feedback coefficients corresponding to the current feature dimension include sub-feedback coefficient 1, sub-feedback coefficient 2 and sub-feedback coefficient 3, which correspond to time window 1, time window 2 and time window 3 respectively. If sub-feedback coefficient 1 and sub-feedback coefficient 2 satisfy the condition that they are less than or equal to the feedback coefficient threshold, then time window 1 and time window 2 are used as target time windows.

[0068] Step S203: The ratio of the number of target time windows to the total number of windows of each time window is used as an active selection factor of the current feature dimension.

[0069] After determining the target time window, the number of target time windows can be counted, and the ratio of the number of target time windows to the total number of windows of each time window can be used as the active selection factor of the current feature dimension. For example, if time window 1 and time window 2 are the target time windows, the number of target time windows is 2, and the total number of time windows is 3, then the active selection factor of the current feature dimension is 2 / 3.

[0070] Taking 0.3 as the feedback coefficient threshold as an example, the active selection factor of the current feature dimension i is It can be calculated by the following formula:

[0071]

[0072] in, It indicates that the current feature dimension i corresponds to the current sub-feedback coefficient of time window j, C is a counting operation, which can characterize the number of target time windows, and n represents the total number of time windows.

[0073] In this embodiment, the pre-set feedback coefficient threshold and the current sub-feedback coefficient of the current feature dimension corresponding to each time window can also be used to screen out the target time window, and further obtain the active selection factor of the current feature dimension. In this way, the active selection factor can be used to characterize whether the number of features within the threshold range of each feature dimension is sufficient and whether the quality is high enough, thereby improving the accuracy of obtaining the active selection factor.

[0074] Furthermore, if Figure 3 As shown, step S102 may further include:

[0075] Step S301, input the graph convolution features of the current feature dimension into the fully connected layer of the range extender temperature prediction model, and obtain the initial predicted range extender temperature data corresponding to each time window of the current feature dimension through the fully connected layer.

[0076] The fully connected layer is the output layer of the range extender temperature prediction model used to output the predicted range extender temperature data. The initial predicted range extender temperature data is the range extender temperature data predicted by the graph convolution features of the current feature dimension. Since the number of time windows is multiple, the initial predicted range extender temperature data predicted by the current feature dimension can also be multiple, corresponding to different time windows.

[0077] Specifically, the vehicle controller can input the graph convolution features of the current feature dimension into the range extender temperature prediction model, and the range extender temperature prediction model outputs the initial predicted range extender temperature data of each time window under the current feature dimension.

[0078] Step S302 , obtaining the difference between the initial predicted range extender temperature data of each time window and the sample range extender temperature data of each time window, and normalizing the difference of each time window to obtain each current sub-feedback coefficient.

[0079] Afterwards, the error between the initial predicted range extender temperature data and the sample range extender temperature data in each time window can be calculated respectively to obtain the error of each time window in the current feature dimension, and then the above error is normalized to obtain each current sub-feedback coefficient.

[0080] For example, the feedback coefficient of the current feature dimension i can be expressed by the following formula:

[0081]

[0082] in, Represents the output data after the current dimension feature passes through the fully connected layer, represents the n initial predicted range extender temperature data predicted by n time windows under the current feature dimension, y is the actual temperature label, is the sample range extender temperature data of n time windows. Then each difference is normalized to obtain n feedback coefficients between (0, 1), that is, the current sub-feedback coefficients under n different time windows are obtained.

[0083] In this embodiment, the errors between the initial predicted range extender temperature data and the sample range extender temperature data in each time window can be calculated respectively, and the errors can be normalized to obtain each current sub-feedback coefficient. In this way, the accuracy of obtaining the current sub-feedback coefficient can be improved.

[0084] In one embodiment, step S104 may further include: obtaining the fusion weight of each feature dimension according to the active selection factor of each feature dimension; wherein the size of the fusion weight is positively correlated with the size of the active selection factor; and using the fusion weight of each feature dimension, weighted processing is performed on the graph convolution features of each feature dimension to obtain a fused feature.

[0085] Among them, the fusion weight refers to the feature weight used to fuse the graph convolution features output by each feature dimension, and the size of the feature weight is positively correlated with the size of the active selection factor, that is, the larger the active selection factor of a feature dimension, the larger the fusion weight of the feature dimension will be. In this embodiment, the active selection factor can characterize the number and quality of features of the feature dimension within a certain feedback coefficient threshold range. If the active selection factor is larger, then the number of features representing the feature dimension within a certain feedback coefficient threshold range is greater and the quality is higher, so the weight of the corresponding dimension feature needs to be increased, otherwise the weight of the corresponding dimension should be reduced.

[0086] There are many ways to set the fusion weight. For example, the active selection factor itself can be used as the fusion weight, or a certain benchmark value can be added to the active selection factor as the fusion weight, or the active selection factor can be increased or decreased by a set multiple and used as the fusion weight. It only needs to satisfy that the size of the fusion weight is positively correlated with the size of the active selection factor.

[0087] Specifically, after obtaining the active selection factor of each feature dimension, the vehicle controller can use the active selection factor to calculate the fusion weight of the corresponding feature dimension, and then use the feature weights to weight the graph convolution features of each feature dimension to obtain the fused feature.

[0088] Taking the active selection factor as the fusion weight directly and the number of feature dimensions as 3 as an example, the fusion feature It can be obtained through the following formula:

[0089]

[0090] in, They represent the active selection factors of feature dimension 1, feature dimension 2 and feature dimension 3 respectively, and They represent the graph convolution features of feature dimension 1, feature dimension 2, and feature dimension 3 respectively.

[0091] In this embodiment, the fusion weight can also be obtained by actively selecting factors, so that the graph convolution features can be weighted fused using the fusion weight. In this way, the weight of feature fusion can be dynamically adjusted, thereby improving the flexibility and adaptability of the model to features of different dimensions.

[0092] Furthermore, step S103 may further include: obtaining the target feature dimension, and taking the feedback coefficient corresponding to the target feature dimension as the optimal feedback coefficient; the active selection factor corresponding to the target feature dimension is the maximum value of each active selection factor; step S104 may further include: normalizing the optimal feedback coefficient, and using the normalized optimal feedback coefficient to enhance the fusion feature to obtain a time-enhanced feature.

[0093] The target feature dimension refers to the feature dimension corresponding to the maximum value of the active selection factor, and the optimal feedback coefficient is the feedback coefficient of the target feature dimension. In this embodiment, the optimal feedback coefficient can also be determined by the active selection factors of each feature dimension. For example, the largest active selection factor can be found from the active selection factors of each feature dimension, and the feedback coefficient of the feature dimension corresponding to the active selection factor is used as the optimal feedback coefficient. For example, Represent the active selection factors of feature dimension 1, feature dimension 2 and feature dimension 3 respectively. If , then feature dimension 2 is used as the target feature dimension, and the feedback coefficient of feature dimension 2 is used as the optimal feedback coefficient.

[0094] After obtaining the optimal feedback coefficient, the optimal feedback coefficient can be normalized first, and the normalized optimal feedback coefficient can be used to enhance the fusion feature, so as to obtain the time enhancement feature.

[0095] For example, the time-enhanced feature The calculation process is as follows:

[0096]

[0097] in, represents the optimal feedback coefficient, then represents the optimal feedback coefficient after normalization, and It represents the fusion feature.

[0098] In this embodiment, the feedback coefficient of the feature dimension corresponding to the largest active selection factor can be used as the optimal feedback coefficient, and the fusion feature can be enhanced by normalizing the optimal feedback coefficient. This method can further improve the accuracy of obtaining the time enhancement feature.

[0099] In addition, the fusion feature is composed of multiple sub-fusion features, each of which corresponds to a different time window; the time enhancement feature is composed of multiple sub-time enhancement features, each of which corresponds to a different time window; Figure 4 As shown, step S105 may further include:

[0100] Step S401, obtaining the sub-time enhancement feature corresponding to the current time window, and the sub-time enhancement feature and sub-fusion feature of the previous time window of the current time window; the current time window is any one of the time windows.

[0101] In this embodiment, the fusion feature can also be composed of multiple sub-fusion features, each sub-fusion feature corresponds to a different time window. Similarly, the time enhancement feature can also be composed of multiple sub-time enhancement features, and each sub-time enhancement feature also corresponds to a different time window, and the current time window can refer to any one of the time windows.

[0102] Specifically, when predicting the range extender temperature, the current time window can also be determined from multiple time windows to obtain the sub-time enhancement features of the current time window, the sub-time enhancement features of the previous time window of the current time window, and the sub-fusion features without feature enhancement.

[0103] Step S402, using the sub-time enhancement feature corresponding to the current time window, and the sub-time enhancement feature and sub-fusion feature of the previous time window, to obtain the temporal enhancement feature corresponding to the current time window.

[0104] After obtaining the sub-time enhancement features corresponding to the current time window, as well as the sub-time enhancement features and sub-fusion features of the previous time window, the above features can be used to obtain the timing enhancement features corresponding to the current time window, and the timing enhancement features can be used to predict the range extender temperature data of the next time window of the current time window.

[0105] For example, the current time window Timing enhancement features It can be represented by the following formula:

[0106]

[0107] in, Indicates the current time window The time enhancement feature under is the weight matrix, represents the previous time window of the current time window, that is, The time enhancement feature under represents the previous time window of the current time window, that is, The fusion features without temporal enhancement under is the bias matrix.

[0108] Step S403, inputting the time series enhancement features into the fully connected layer of the range extender temperature prediction model, and obtaining the predicted range extender temperature data of the next time window of the current time window through the fully connected layer.

[0109] After obtaining the timing enhancement features corresponding to the current time window, the timing enhancement features can be input into the fully connected layer used to output the predicted range extender temperature data in the range extender temperature prediction model, and the predicted range extender temperature data of the next time window of the current time window can be obtained through the fully connected layer.

[0110] In this embodiment, the time series enhancement feature corresponding to the current time window can be obtained by combining the sub-time enhancement feature and the sub-fusion feature of the previous time window to output the predicted range extender temperature data of the next time window of the current time window. In this way, a recursive feature selection mechanism can be implemented, so that the model can dynamically adjust the importance of the features according to the feedback information, so as to better adapt to the changes and dynamics of the data and improve the adaptability of the model.

[0111] In one embodiment, Figure 5 As shown, step S101 may further include:

[0112] Step S501: taking each time window as a graph node of a sample range extender temperature characteristic graph, and taking the sample range extender temperature data of each time window as a node feature of each graph node.

[0113] In this embodiment, the graph nodes of the sample range extender temperature characteristic graph can represent different time windows of the sample range extender temperature data, and the sample range extender temperature data of each time window can be used as the node feature of each graph node. For example, the sample range extender temperature data of each time window can contain 1,000 data samples, and the above data samples can be used as the node feature of each graph node.

[0114] Step S502, using the features of each node, obtaining the similarity between each graph node, and obtaining an adjacency matrix for characterizing the connection relationship between each graph node according to the similarity;

[0115] Step S503: construct a temperature characteristic graph of a sample range extender according to the adjacency matrix and the characteristics of each node.

[0116] After that, the node features of each graph node can be used to find the similarity between the graph nodes. For example, the cosine similarity can be used to calculate the adjacency matrix, which can characterize the connection relationship between the graph nodes. Taking n time windows as an example, the number of graph nodes is n, and the adjacency matrix obtained at this time can be an n×n adjacency matrix. After that, the sample range extender temperature characteristic graph can be constructed based on the constructed adjacency matrix and the features of each node.

[0117] For example, the graph structure can be represented by G=(A,X), where X is the node feature, i.e., the sample range extender temperature data of each time window, and A represents the adjacency matrix of the graph G, representing the connection relationship between the nodes, which can be obtained by calculating the similarity between the nodes of each graph.

[0118] In this embodiment, the sample range extender temperature data of each time window can also be used as the node features of the graph nodes, and the similarity between the graph nodes can be obtained through the node features to construct an adjacency matrix to form a sample range extender temperature characteristic graph. In this way, the efficiency of constructing the sample range extender temperature characteristic graph can be improved.

[0119] In one embodiment, Figure 6 As shown, a range extender control method is also provided. This embodiment uses the method applied to a vehicle controller as an example. In this embodiment, the method includes the following steps:

[0120] Step S601, obtaining historical range extender temperature data corresponding to different time windows of a target range extender, and constructing a historical range extender temperature characteristic diagram based on the historical range extender temperature data of each time window.

[0121] Among them, the target range extender refers to the range extender for which range extender temperature prediction is required, and the historical range extender temperature data refers to the temperature data of the target range extender in different time windows, and the historical range extender temperature characteristic graph is a characteristic graph constructed based on the above temperature data.

[0122] Specifically, when it is necessary to predict the temperature of the target range extender, the historical range extender temperature data corresponding to different time windows of the range extender can be collected first, and then the corresponding historical range extender temperature characteristic diagram can be constructed based on the above historical range extender temperature data.

[0123] Step S602, inputting the historical range extender temperature characteristic diagram into the trained range extender temperature prediction model, and outputting the predicted range extender temperature data of the target range extender through the range extender temperature prediction model; wherein the range extender temperature prediction model is trained by the range extender temperature prediction model training method of any of the above embodiments.

[0124] Afterwards, the vehicle controller can input the historical range extender temperature characteristic diagram into the trained range extender temperature prediction model, and the range extender temperature prediction model outputs the predicted range extender temperature data of the target range extender.

[0125] Among them, the predicted range extender temperature data can be obtained through the following process: first, the range extender temperature prediction model extracts the graph convolution features corresponding to different feature dimensions of the historical range extender temperature characteristic graph, and obtains the feedback coefficient of each graph convolution feature, which can be input into the fully connected layer of the range extender temperature prediction model, and after obtaining the initial prediction data of each time window, the feedback coefficient of each graph convolution feature is obtained based on the error between the initial prediction data and the historical range extender temperature data. Then, according to the above feedback coefficient, the active selection factor and the optimal feedback coefficient of each feature dimension are obtained, which can be first compared with each feedback coefficient and the preset feedback coefficient threshold, so as to count the number of time windows in each feedback coefficient that is less than or equal to the feedback coefficient threshold, to calculate the active selection factor of each feature dimension, and further take the feedback coefficient of the feature dimension with the largest active selection factor as the optimal feedback coefficient. Next, the active selection factor can be used to fuse the various graph convolution features, and the fused features can be enhanced based on the optimal feedback coefficient. The active selection factor can be used to obtain the fusion weight for weighted fusion, and the optimal feedback coefficient can be normalized to enhance the fusion weight using the normalized optimal feedback coefficient. Finally, the time enhancement feature is used to obtain the predicted range extender temperature data. For example, the time enhancement feature and fusion feature of the previous time interval of the current time interval and the time enhancement feature of the current time interval can be used to output the predicted range extender temperature data of the next time window.

[0126] Step S603: temperature control the target range extender using the predicted range extender temperature data.

[0127] In the above-mentioned range extender control method, by acquiring the historical range extender temperature data of the target range extender corresponding to different time windows, and based on the historical range extender temperature data of each time window, a historical range extender temperature characteristic graph is constructed; the historical range extender temperature characteristic graph is input into the trained range extender temperature prediction model, and the predicted range extender temperature data of the target range extender is output through the range extender temperature prediction model; wherein the range extender temperature prediction model is trained by the range extender temperature prediction model training method as described in any of the above embodiments; and the predicted range extender temperature data is used to control the temperature of the target range extender. The present invention constructs a historical range extender temperature characteristic diagram of a target range extender, and outputs predicted range extender temperature data in combination with a trained range extender temperature prediction model. Since multi-dimensional feature output and fusion are achieved during range extender temperature prediction model training, feature representation of data can be learned more comprehensively, and active selection factors can be obtained based on feedback coefficients of each feature dimension, and feature fusion can be performed based on the active selection factors. Feature enhancement can also be performed using the optimal feedback coefficient. The trained range extender temperature prediction model can accurately predict the range extender temperature, thereby improving the accuracy of range extender temperature prediction and further improving the stability of range extender control.

[0128] In one embodiment, a range extender temperature prediction method based on active sample selective enhancement is also provided. Figure 7 As shown in the figure, this method designs a multi-dimensional feature enhancement mechanism to adaptively select the feature output of the optimal dimension for different working condition data. Secondly, a recursive feature selection mechanism is designed to effectively enhance the time series feature extraction based on the feedback coefficient, improve the generalization ability and adaptability of the prediction model under different working conditions, and improve the accuracy of the prediction. Specifically, it includes the following steps:

[0129] First, the historical temperature data of the range extender under multiple working conditions are collected, and then a multi-dimensional feature enhancement mechanism is designed to divide the historical temperature data into time windows T, T=[t 1 , t 2 ,…, t n], where n represents the nth time window, and each time window contains 1000 data samples. First, we need to construct a graph for the historical temperature of the range extender. The graph structure is the input of the graph convolutional network. The graph structure consists of nodes, features corresponding to the nodes, and connecting lines between nodes. The graph structure is represented by G=(A,X), where X is the node feature and A represents the adjacency matrix of the graph G, which represents the connection relationship between the nodes. Therefore, each time window is regarded as a node, and the 1000 data samples in the time window are used as the node features corresponding to the node, so that n nodes corresponding to n time windows and 1000 data samples in each time window can be obtained as the node features of the corresponding nodes. Then, the similarity between the nodes is calculated by cosine similarity, so that an n×n adjacency matrix can be obtained, that is, A and X in the above graph structure are node features. Then, the graph structure is subjected to graph convolution operation to mine the hidden spatial feature information between time windows, and construct feature outputs of three dimensions to capture the features of the data from different angles, so that the model can learn the feature representation of the data more comprehensively. The calculation process is as follows:

[0130]

[0131]

[0132]

[0133] in The output feature dimension is 200. The output feature dimension is 400. The output feature dimension is 600. , , are the corresponding weight matrices, To perform symmetric normalization on the adjacency matrix A, is the activation function. In this way, we can obtain n nodes with node features of 200, 400, and 600, which can be expressed as n×200, n×400, and n×600. Then, the three dimensional features are input into the fully connected layer for preliminary prediction, and the feedback coefficient of each dimensional output feature is calculated. The calculation process is as follows:

[0134]

[0135] Among them, i=1, 2, 3, respectively represent , , Three dimensional features: Represents the initial temperature prediction value obtained after the three dimensional features pass through the fully connected layer, represents the n temperature prediction values ​​obtained by n nodes under the i-th dimension, y is the actual temperature label, is n actual temperature values. Then the difference is normalized to obtain 3×n feedback coefficients between (0, 1).

[0136] Based on the feedback coefficient, an active selection factor is constructed. The feedback coefficients of the three dimensions are proportionally calculated respectively. By setting the threshold, the number of nodes that meet the threshold is calculated to obtain the fusion weights corresponding to the features of different dimensions. The calculation process is as follows:

[0137]

[0138] Among them, i=1, 2, 3 correspond to three dimensions. It represents the feedback coefficient of the jth initial temperature prediction value predicted by the i-th feature dimension, 0.3 is the threshold, and C is the counting operation. In this way, the active selection factor under each feature dimension can be obtained. Through this proportional calculation, a high Z value represents a large number of features within the threshold range and high quality, so the weight of the corresponding dimension feature needs to be increased, and vice versa, the weight of the corresponding dimension should be reduced. Therefore, the active selection factor is given to dynamically adjust the weight of feature fusion, which improves the flexibility and adaptability of the model to features of different dimensions. The multi-dimensional feature fusion process is as follows:

[0139]

[0140] Secondly, a recursive feature selection mechanism is designed. Since the above operation is mainly used to mine the spatial correlation information between time series signals without considering the temporal correlation, the features at each moment are first enhanced based on the feedback coefficient. First, the feedback coefficient under the optimal dimension is extracted, that is, The corresponding optimal feedback coefficient is Indicates that Normalization is performed, that is, ,because The smaller the prediction, the more accurate it is. Multiply it with the feature at each moment to achieve time feature enhancement. The corresponding calculation process of the time enhancement feature is as follows:

[0141]

[0142] Then, an effective sample selection factor is designed to select sample features from the features of the previous moment. In this process, the features of the unenhanced moment are retained. The effective sample selection factor is used to adaptively select features to ensure that the model can comprehensively utilize all feature information during the entire training process and improve the global performance of the model. The calculation process is as follows:

[0143]

[0144] in, represent Time enhancement features under time window, express Features without temporal enhancement under the time window, It represents the temporal enhancement feature obtained by mining multiple feature information of the previous moment. is the bias matrix. Finally, the feature The input is sent to the fully connected layer, and the predicted temperature of the range extender in the next time period is output, thereby achieving the temperature prediction task.

[0145] Through this embodiment, a range extender temperature prediction method based on active sample selective enhancement is proposed to solve the problem that the existing range extender temperature prediction model has high requirements on data sample quality and the model is complex, which limits the accuracy and stability of the model in temperature prediction. The main method includes: first, the historical temperature data of the range extender under multiple working conditions is collected, and then a multi-dimensional feature enhancement mechanism is designed. The historical temperature data is divided according to time windows, and each time window contains 1,000 data samples. The spatial feature information hidden between the time windows is mined through graph convolution feature aggregation, and three dimensions of feature output are constructed to capture the features of the data from different angles, so that the model can learn the feature representation of the data more comprehensively. Then, the feedback coefficient of the output feature of each dimension is calculated, and then the active selection factor is constructed to obtain the fusion weights corresponding to the features of different dimensions. In this way, the weights of feature fusion are dynamically adjusted, which improves the flexibility and adaptability of the model to features of different dimensions. Then, features of different dimensions are fused according to the fusion weights to obtain multi-dimensional fusion features; secondly, a recursive feature selection mechanism is designed to enhance the samples of the features of the previous moment based on the feedback coefficient and the effective sample selection factor. The model can dynamically adjust the importance of the features according to the feedback information, so as to better adapt to the changes and dynamics of the data and improve the adaptability of the model.

[0146] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0147] Based on the same inventive concept, the embodiment of the present invention also provides a range extender temperature prediction model training device for implementing the range extender temperature prediction model training method involved above, and a range extender temperature prediction device for implementing the range extender temperature prediction method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more range extender temperature prediction model training device embodiments and range extender temperature prediction device embodiments provided below can refer to the limitations of the range extender temperature prediction model training method and the range extender temperature prediction method above, and will not be repeated here.

[0148] In one embodiment, Figure 8 As shown, a range extender temperature prediction model training device is provided, comprising: a sample characteristic diagram construction module 801, a feedback coefficient acquisition module 802, a selection factor acquisition module 803, an enhanced feature acquisition module 804 and a prediction model training module 805, wherein:

[0149] The sample characteristic diagram construction module 801 is used to obtain sample range extender temperature data corresponding to different time windows under multiple working conditions, and to construct a sample range extender temperature characteristic diagram based on the sample range extender temperature data of each time window;

[0150] A feedback coefficient acquisition module 802 is used to input the sample range extender temperature characteristic graph into the range extender temperature prediction model to be trained, extract graph convolution features corresponding to different feature dimensions of the sample range extender temperature characteristic graph through the range extender temperature prediction model, and acquire feedback coefficients of each feature dimension based on each graph convolution feature;

[0151] A selection factor acquisition module 803 is used to acquire active selection factors of each feature dimension according to each feedback coefficient, and to acquire an optimal feedback coefficient from each feedback coefficient;

[0152] The enhanced feature acquisition module 804 is used to fuse the graph convolution features of each feature dimension using the active selection factor of each feature dimension to obtain a fused feature, and to enhance the fused feature based on the optimal feedback coefficient to obtain a time-enhanced feature;

[0153] The prediction model training module 805 is used to obtain the predicted range extender temperature data based on the time enhancement feature, and train the range extender temperature prediction model using the predicted range extender temperature data to obtain a trained range extender temperature prediction model.

[0154] In one embodiment, the feedback coefficient of each feature dimension is composed of multiple sub-feedback coefficients, and each sub-feedback coefficient corresponds to a different time window; the selection factor acquisition module 803 is further used to obtain the current feature dimension and the current sub-feedback coefficient of the current feature dimension corresponding to each time window; the current feature dimension is any one of the feature dimensions; according to the pre-set feedback coefficient threshold and each current sub-feedback coefficient, the target time window corresponding to the current feature dimension and the number of target time windows are obtained from each time window; the current sub-feedback coefficient corresponding to the target time window is less than or equal to the feedback coefficient threshold; the ratio of the number of target time windows to the total number of windows of each time window is used as the active selection factor of the current feature dimension.

[0155] In one embodiment, the feedback coefficient acquisition module 802 is further used to input the graph convolution features of the current feature dimension into the fully connected layer of the range extender temperature prediction model, and obtain the initial predicted range extender temperature data corresponding to each time window of the current feature dimension through the fully connected layer; obtain the initial predicted range extender temperature data of each time window, and the difference between the sample range extender temperature data of each time window, and normalize the difference of each time window to obtain each current sub-feedback coefficient.

[0156] In one embodiment, the enhanced feature acquisition module 804 is further used to obtain the fusion weight of each feature dimension according to the active selection factor of each feature dimension; wherein the size of the fusion weight is positively correlated with the size of the active selection factor; and the graph convolution features of each feature dimension are weightedly processed using the fusion weight of each feature dimension to obtain the fused feature.

[0157] In one embodiment, the selection factor acquisition module 803 is further used to acquire the target feature dimension, and use the feedback coefficient corresponding to the target feature dimension as the optimal feedback coefficient; the active selection factor corresponding to the target feature dimension is the maximum value of each active selection factor; the enhanced feature acquisition module 804 is further used to normalize the optimal feedback coefficient, and use the normalized optimal feedback coefficient to enhance the fusion feature to obtain the time enhanced feature.

[0158] In one embodiment, the fusion feature is composed of multiple sub-fusion features, each sub-fusion feature corresponds to a different time window; the time enhancement feature is composed of multiple sub-time enhancement features, each sub-time enhancement feature corresponds to a different time window; the prediction model training module 805 is further used to obtain the sub-time enhancement feature corresponding to the current time window, as well as the sub-time enhancement feature and sub-fusion feature of the previous time window of the current time window; the current time window is any one of the time windows; the sub-time enhancement feature corresponding to the current time window, as well as the sub-time enhancement feature and sub-fusion feature of the previous time window, are used to obtain the timing enhancement feature corresponding to the current time window; the timing enhancement feature is input into the fully connected layer of the range extender temperature prediction model, and the predicted range extender temperature data of the next time window of the current time window is obtained through the fully connected layer.

[0159] In one embodiment, the sample feature graph construction module 801 is further used to use each time window as a graph node of the sample range extender temperature feature graph, and use the sample range extender temperature data of each time window as the node feature of each graph node; using the features of each node, the similarity between each graph node is obtained, and based on the similarity, an adjacency matrix for characterizing the connection relationship between each graph node is obtained; based on the adjacency matrix and the features of each node, a sample range extender temperature feature graph is constructed.

[0160] In one embodiment, Fig. 9 As shown, a range extender control device is provided, including: a historical characteristic map construction module 901, a predicted temperature output module 902 and a range extender temperature control module 903, wherein:

[0161] A historical characteristic map construction module, used to obtain historical range extender temperature data corresponding to different time windows of the target range extender, and to construct a historical range extender temperature characteristic map based on the historical range extender temperature data of each time window;

[0162] A predicted temperature output module 902 is used to input the historical range extender temperature characteristic diagram into the trained range extender temperature prediction model, and output the predicted range extender temperature data of the target range extender through the range extender temperature prediction model; wherein the range extender temperature prediction model is trained by the range extender temperature prediction model training method of any of the above embodiments;

[0163] The range extender temperature control module 903 is used to perform temperature control on the target range extender using the predicted range extender temperature data.

[0164] The above-mentioned range extender temperature prediction model training device and each module in the range extender temperature prediction device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the vehicle controller in the form of hardware, or can be stored in the memory in the vehicle controller in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.

[0165] In one embodiment, a vehicle controller is provided. The vehicle controller may be a terminal, and its internal structure diagram may be as follows: Fig.10 As shown. The vehicle controller includes a processor, a memory, an input / output interface and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the vehicle controller is used to provide computing and control capabilities. The memory of the vehicle controller includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the vehicle controller is used to exchange information between the processor and an external device. The communication interface of the vehicle controller is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a range extender temperature prediction model training method and a range extender temperature prediction method are implemented.

[0166] Those skilled in the art will understand that Fig.10 The structure shown in the figure is merely a block diagram of a partial structure related to the scheme of the present invention, and does not constitute a limitation on the vehicle controller to which the scheme of the present invention is applied. The specific vehicle controller may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0167] In one embodiment, a vehicle controller is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0168] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0169] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0170] 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 the present invention are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0171] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided by the present invention may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided by the present invention may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0172] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0173] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the attached claims.

Claims

1. A range extender temperature prediction model training method, characterized in that: The method comprises: Acquire sample range extender temperature data corresponding to different time windows under multiple operating conditions, and construct a sample range extender temperature characteristic diagram based on the sample range extender temperature data of each of the time windows; Inputting the sample range extender temperature characteristic graph into the range extender temperature prediction model to be trained, extracting graph convolution features of the sample range extender temperature characteristic graph corresponding to different feature dimensions through the range extender temperature prediction model, and obtaining feedback coefficients of each feature dimension based on each graph convolution feature; According to each of the feedback coefficients, an active selection factor of each of the feature dimensions is obtained, and an optimal feedback coefficient is obtained from each of the feedback coefficients; Using the active selection factor of each feature dimension, the graph convolution features of each feature dimension are fused to obtain a fused feature, and the fused feature is enhanced based on the optimal feedback coefficient to obtain a time-enhanced feature; Predicted range extender temperature data is acquired based on the time enhancement feature, and the range extender temperature prediction model is trained using the predicted range extender temperature data to obtain a trained range extender temperature prediction model.

2. The method according to claim 1, characterized in that The feedback coefficient of each feature dimension is composed of a plurality of sub-feedback coefficients, each of which corresponds to a different time window; and obtaining the active selection factor of each feature dimension according to each feedback coefficient includes: Acquire a current feature dimension, and a current sub-feedback coefficient of each time window corresponding to the current feature dimension; the current feature dimension is any one of the feature dimensions; According to a preset feedback coefficient threshold and each of the current sub-feedback coefficients, a target time window corresponding to the current feature dimension and the number of the target time windows are obtained from each of the time windows; the current sub-feedback coefficient corresponding to the target time window is less than or equal to the feedback coefficient threshold; The ratio of the number of the target time windows to the total number of the time windows is used as the active selection factor of the current feature dimension.

3. The method according to claim 2, characterized in that The obtaining of feedback coefficients of the feature dimensions based on the graph convolution features includes: Inputting the graph convolution feature of the current feature dimension into the fully connected layer of the range extender temperature prediction model, and obtaining the initial predicted range extender temperature data corresponding to each time window of the current feature dimension through the fully connected layer; The initial predicted range extender temperature data of each time window and the difference between the sample range extender temperature data of each time window are obtained, and the difference of each time window is normalized to obtain each current sub-feedback coefficient.

4. The method according to claim 1, characterized in that: The active selection factor of each feature dimension is used to fuse the graph convolution features of each feature dimension to obtain the fused features, including: According to the active selection factor of each feature dimension, a fusion weight of each feature dimension is obtained; wherein the size of the fusion weight is positively correlated with the size of the active selection factor; The graph convolution features of each feature dimension are weighted by utilizing the fusion weights of the feature dimensions to obtain the fused features.

5. The method according to claim 4, characterized in that The obtaining of the optimal feedback coefficient from the feedback coefficients comprises: Acquire a target feature dimension, and use the feedback coefficient corresponding to the target feature dimension as the optimal feedback coefficient; the active selection factor corresponding to the target feature dimension is the maximum value of each of the active selection factors; The step of enhancing the fusion feature based on the optimal feedback coefficient to obtain a time-enhanced feature includes: The optimal feedback coefficient is normalized, and the fusion feature is enhanced using the normalized optimal feedback coefficient to obtain the time-enhanced feature.

6. The method according to claim 1, characterized in that The fusion feature is composed of a plurality of sub-fusion features, each of which corresponds to a different time window; the time enhancement feature is composed of a plurality of sub-time enhancement features, each of which corresponds to a different time window; the method of obtaining the predicted range extender temperature data based on the time enhancement feature includes: Obtaining the sub-time enhancement feature corresponding to the current time window, and the sub-time enhancement feature and the sub-fusion feature of the previous time window of the current time window; the current time window is any one of the time windows; Obtaining a time series enhancement feature corresponding to the current time window by using the sub-time enhancement feature corresponding to the current time window, and the sub-time enhancement feature and the sub-fusion feature of the previous time window; The timing enhancement feature is input into the fully connected layer of the range extender temperature prediction model, and the predicted range extender temperature data of the next time window of the current time window is obtained through the fully connected layer.

7. The method according to any one of claims 1 to 6, characterized in that: The step of constructing a sample range extender temperature characteristic diagram based on the sample range extender temperature data of each time window includes: Using each of the time windows as a graph node of the sample range extender temperature characteristic graph, and using the sample range extender temperature data of each of the time windows as a node feature of each of the graph nodes; Using the node features, the similarity between the graph nodes is obtained, and based on the similarity, an adjacency matrix is ​​obtained for characterizing the connection relationship between the graph nodes. The sample range extender temperature characteristic graph is constructed according to the adjacency matrix and each of the node characteristics.

8. A range extender control method, characterized in that: The method comprises: Acquire historical range extender temperature data corresponding to different time windows of the target range extender, and construct a historical range extender temperature characteristic diagram based on the historical range extender temperature data of each of the time windows; Inputting the historical range extender temperature characteristic diagram into a trained range extender temperature prediction model, and outputting predicted range extender temperature data of the target range extender through the range extender temperature prediction model; wherein the range extender temperature prediction model is trained by the range extender temperature prediction model training method according to any one of claims 1 to 7; The target range extender is temperature controlled using the predicted range extender temperature data.

9. A vehicle controller, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

11. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images