Electric vehicle connector temperature rise prediction method based on LHS optimization GA-BP algorithm

Abstract

The invention relates to an electric vehicle connector temperature rise prediction method based on an LHS optimization GA-BP algorithm, and the method comprises the steps: obtaining an optimization instruction, building an optimization data set based on a preset sampling density and a preset LHS sampling algorithm, and carrying out the optimization of a pre-built BP neural network model through the optimization data set and a preset GA algorithm; when a prediction instruction is received, inputting a thermal simulation parameter value contained in the prediction instruction into the BP neural network model and outputting a prediction result; based on a response output result of the BP neural network model to each data interval in an optimization process, determining first response sensitivity and sampling density of each data interval, defining multi-order sampling density for each thermal simulation parameter of the high-sensitivity interval, and sampling the thermal simulation parameters in the high-sensitivity interval step by step according to the corresponding multi-order sampling density to obtain a high-sensitivity thermal simulation parameter; and determining the second response sensitivity and the sampling density of each thermal simulation parameter in the high-sensitivity interval, and updating the sampling density. The method has the effect of improving the temperature rise prediction precision and stability.

Description

Technical Field

[0001] The present application relates to the technical field of temperature rise prediction of electric vehicle connectors, and in particular to a method for predicting temperature rise of electric vehicle connectors based on an LHS optimized GA-BP algorithm. Background Art

[0002] In recent years, China's electric vehicle industry has achieved rapid growth, but range anxiety remains a major concern for users, which has largely restricted the further expansion of the electric vehicle market. DC charging technology can provide electric vehicles with rapid energy replenishment, which is expected to significantly alleviate users' range anxiety. In the DC charging system, the DC charging connector is not only a bridge between the power supply equipment and the electric vehicle, but also a weak link in the entire charging system and a key factor in the connectivity and safety of DC charging.

[0003] In the context of the rapid development of the current electric vehicle industry, in order to meet the urgent demand for temperature rise suppression technology, the industry has taken a variety of measures, including changing the connection method, optimizing the terminal structure, improving the plating process, and introducing a liquid cooling system. However, these measures involve a wide range of fields and require a long period of testing and verification, which is in conflict with the electric vehicle industry's requirements for rapid product iteration. In order to ensure quality while ensuring development progress, accurate temperature prediction analysis must be performed during the product design phase to shorten the development cycle. It is crucial to develop a high-precision, fast, and adaptable temperature prediction method that can predict performance in the early design stage and reduce time and cost losses caused by design adjustments in the later stage, thereby meeting the industry's demand for rapid response to the market.

[0004] The traditional BP neural network is learned and trained through forward signal transmission and reverse error propagation. The model is prone to fall into the local optimal solution due to the randomness of the initial weights and thresholds, which ultimately leads to limited stability and accuracy of the prediction results, so it needs to be improved. Summary of the invention

[0005] In order to optimize the prediction stability and accuracy of the temperature rise data of electric vehicle connectors, the present application provides an electric vehicle connector temperature rise prediction method based on the LHS optimized GA-BP algorithm.

[0006] In the first aspect, the present application provides a method for predicting the temperature rise of an electric vehicle connector based on an LHS optimized GA-BP algorithm, which adopts the following technical solutions: Obtaining optimization instructions, constructing an optimization data set formed by thermal simulation parameter values ​​of the electric vehicle connector based on a preset sampling density and a preset LHS sampling algorithm; optimizing a pre-constructed BP neural network model using the optimization data set and a preset GA algorithm; Whenever a prediction instruction is received, the thermal simulation parameter value contained in the prediction instruction is used as an input parameter and input into the latest optimized BP neural network model, and the prediction result with temperature data is output through the BP neural network model; Based on the response output results of the BP neural network model to the data set corresponding to each data interval during the optimization process, the first response sensitivity and sampling density of each data interval are determined; wherein the data interval refers to a number of numerical intervals formed by dividing the thermal simulation parameters when sampling by the LHS sampling algorithm; the first response sensitivity is used to characterize the significance of the influence of the input variables of the corresponding data interval on the output results of the BP neural network model, and the more significant the influence, the higher the corresponding response sensitivity and the greater the sampling density; According to the first response sensitivity of each data interval, a multi-order sampling density is defined for each thermal simulation parameter in the high-sensitivity interval, and after sampling the thermal simulation parameters in the high-sensitivity interval step by step according to the corresponding multi-order sampling density, a second response sensitivity and sampling density of each thermal simulation parameter in the high-sensitivity interval are determined; wherein the high-sensitivity interval refers to a data interval in which the first response sensitivity is higher than the preset sensitivity; the second response sensitivity is used to characterize the significance of the influence of the sampling density of the thermal simulation parameters in the high-sensitivity interval on the output result of the BP neural network model; The determined sampling density is used to update the preset sampling density of the non-high-sensitivity interval and the sampling density corresponding to each thermal simulation parameter of the high-sensitivity interval.

[0007] By adopting the above technical solution, the multivariate thermal simulation parameters are sampled based on the Latin hypercube sampling algorithm (LHS algorithm) to ensure sampling uniformity; the sampling data is then used to train the BP neural network model, so that the BP neural network model continuously reversely adjusts the network weights and thresholds through its own learning mechanism to reduce the prediction error, and further, the present application proposes to optimize the BP neural network model using a genetic algorithm (GA algorithm), and finally realize the temperature rise prediction of the electric vehicle connector through the optimized BP neural network model to improve the prediction accuracy and prediction efficiency; in addition, the present application adjusts the sampling density according to the significance of the influence of the thermal simulation parameter values ​​on the prediction results, increases the sampling density for the sample intervals with significant influence, and reduces the sampling density for the sample intervals with insignificant influence, so as to reasonably allocate computing resources, and appropriately reduce unnecessary sampling and training operations on the basis of ensuring the training effect of the BP neural network model.

[0008] Optionally, the method further includes: The charging information is acquired in real time based on the onboard system of the electric vehicle connected in advance, and the charging efficiency of the electric vehicle connector is determined according to the charging information. When the charging efficiency meets the preset trigger condition, the optimization instruction is triggered; Recording the BP neural network model after each optimization and the corresponding charging efficiency, and determining the effective prediction range of the BP neural network model; Whenever a prediction instruction is received, the thermal simulation parameter value contained in the prediction instruction is used as an input parameter and input into the latest optimized BP neural network model, and the prediction result with temperature data is outputted by the BP neural network model, including: Whenever a prediction instruction is received, determining whether the prediction time is within the effective prediction range of the currently optimized BP neural network model according to the prediction time included in the prediction instruction; If so, the thermal simulation parameter values ​​contained in the prediction instruction are used as input parameters and input into the latest optimized BP neural network model, and the prediction results with temperature data are output through the BP neural network model; if not, the prediction results with temperature data are output based on the preset proxy model prediction.

[0009] By adopting the above technical solution, when the same electric vehicle connector is input into the BP neural network model with the same thermal simulation data as the input parameter value to obtain the output result at different stages, the corresponding output results may be different. The reason may be that the electric vehicle connector leads to performance degradation and aging of physical materials as the service life shortens, which in turn affects the temperature rise data in reverse. For this reason, the present application proposes to use charging efficiency to characterize the performance changes of electric vehicle connectors, and construct a corresponding relationship between charging efficiency and BP neural network model, so as to select the corresponding BP neural network model at different stages of electric vehicles (i.e., stages corresponding to different charging efficiencies), thereby realizing adaptive adjustment of temperature rise prediction technology according to performance changes of electric vehicle connectors and ensuring the accuracy of temperature rise prediction.

[0010] Optionally, whenever a prediction instruction is received, determining whether the prediction time is within a valid prediction range of the currently optimized BP neural network model according to the prediction time included in the prediction instruction includes: Obtain the changing trend of charging efficiency over time, and predict the effective period of the model corresponding to the charging efficiency corresponding to the currently optimized BP neural network model, where the effective period of the model refers to the time period for maintaining the corresponding charging efficiency; If the prediction time is within the effective period of the model, it is considered to be within the effective prediction range of the currently optimized BP neural network model; otherwise, it is considered not to be within the effective prediction range of the currently optimized BP neural network model.

[0011] By adopting the above technical solution, combined with the operating status of the electric vehicle connector itself (i.e., the changing trend of charging efficiency over time), the time period during which each charging efficiency can be maintained is predicted, so that the effective time period of the model corresponding to the currently optimized BP neural network model can be determined. The present application believes that after exceeding the effective time period of the model, the temperature data predicted by the BP neural network model will exceed the fault tolerance range, that is, the error is large. Therefore, the present application determines whether the predicted time is within the effective time period of the model as the basis for determining whether the corresponding prediction instruction can be predicted and executed by the currently optimized BP neural network model.

[0012] Optionally, the method further includes: Whenever the BP neural network model is optimized, the temperature deviation of the BP neural network model corresponding to the optimized time is determined, and the pre-built proxy model is trained with the charging efficiency difference as the input parameter and the temperature deviation as the output parameter; if not, the prediction result with temperature data is output based on the preset proxy model prediction, including: If not, the thermal simulation parameter values ​​contained in the prediction instruction are used as input parameters and input into the latest optimized BP neural network model, and the initial temperature data is output through the BP neural network model; Predicting a charging efficiency corresponding to the predicted time, determining a charging efficiency difference between the charging efficiency corresponding to the predicted time and the charging efficiency corresponding to the current BP neural network model, and outputting a temperature deviation based on the charging efficiency difference through a proxy model; The initial temperature data and the temperature deviation are fused to obtain temperature data, and a prediction result including the temperature data is output.

[0013] By adopting the above-mentioned technical scheme, the present application uses a proxy model to predict the prediction error (i.e., temperature deviation) after using the current BP neural network model to predict the prediction instruction that exceeds its effective prediction range, and then merges the prediction result (i.e., initial temperature data) obtained after using the current BP neural network model to predict the prediction instruction that exceeds its effective prediction range with the temperature deviation to finally predict the temperature data. In addition, the present application will regularly train the proxy model with the charging efficiency difference as the input parameter and the temperature deviation as the output parameter through the temperature deviation of the BP neural network model optimized corresponding to different times, so as to improve the prediction accuracy of the proxy model, thereby reducing the deviation between the final predicted temperature data and the actual temperature data, and improving the prediction accuracy.

[0014] Optionally, the thermal simulation parameter value included in the prediction instruction includes at least a charging duration; the prediction time is a charging period corresponding to a combination of a charging start time and a charging duration; If not, the thermal simulation parameter value included in the prediction instruction is used as an input parameter and input into the latest optimized BP neural network model, and the initial temperature data is output through the BP neural network model; the corresponding charging efficiency is predicted according to the prediction time, and the charging efficiency difference between the charging efficiency corresponding to the prediction time and the charging efficiency corresponding to the current BP neural network model is determined, and the temperature deviation is output based on the charging efficiency difference through the proxy model, the initial temperature data and the temperature deviation are merged to obtain temperature data, and the prediction result with the temperature data is output, including: if not, the thermal simulation parameter value included in the prediction instruction is predicted through the current BP neural network model and the initial temperature data is output; Determine whether the prediction time has an intersection with the model effective period corresponding to the current BP neural network model; If there is no intersection, determining a first charging efficiency difference between the first charging efficiency and the charging efficiency corresponding to the current BP neural network model according to the first charging efficiency predicted corresponding to the predicted time, outputting a first temperature deviation based on the first charging efficiency difference through a proxy model, fusing the initial temperature data with the first temperature deviation to obtain temperature data, and outputting a prediction result with the temperature data; If there is an intersection, the time period corresponding to the intersection in the forecast time is taken as the effective forecast period, and the time period of the non-effective forecast period is taken as the deviation forecast period; The initial temperature data is used as the temperature data of the effective prediction period; the second charging efficiency predicted corresponding to the deviation prediction period and the second charging efficiency difference between the second charging efficiency and the charging efficiency corresponding to the current BP neural network model are determined, and the second temperature deviation is output based on the second charging efficiency difference through the proxy model, and the initial temperature data and the second temperature deviation are merged to obtain the temperature data corresponding to the deviation prediction period; the prediction result with the effective prediction period, the deviation prediction period and its corresponding Wendy data is output.

[0015] By adopting the above technical scheme, this scheme takes into account the situation that the charging efficiency changes suddenly during the charging process and triggers the optimization instruction, thereby causing the current BP neural network model to suddenly increase the prediction error of the temperature rise prediction from the time the mutation occurs to the time the charging ends. The prediction time is segmented and analyzed based on the charging time to achieve segmented prediction and minimize the prediction error.

[0016] Optionally, when the charging efficiency meets a preset trigger condition, triggering the optimization instruction includes: Based on the charging efficiency acquired in real time, when the charging efficiency changes and the change is not a transient change, it is determined whether the change meets a preset trigger condition, and if so, an optimization instruction is triggered.

[0017] By adopting the above technical solution, the electric vehicle connector is prone to a sudden change in charging efficiency due to unstable connection, which may cause a brief interruption of charging, etc. during the charging process. However, this change generally disappears after the connection is stabilized again, that is, this change is a short-term change and does not represent a change in the performance of the electric vehicle connector. For this reason, this application excludes the aforementioned situation where there is a short-term change before analyzing whether the change in charging efficiency meets the preset trigger conditions.

[0018] Optionally, determining the second response sensitivity and sampling density of each thermal simulation parameter in the high-sensitivity interval includes: All thermal simulation parameters in the high-sensitivity interval are randomly combined to form a parameter combination, and each of the parameter combinations in the high-sensitivity interval is sampled step by step according to the corresponding multi-order sampling density to determine the third response sensitivity of each of the parameter combinations; wherein the third response sensitivity is used to characterize the significance of the influence of the sampling density of the parameter combination in the high-sensitivity interval on the output result of the BP neural network model; Based on the second response sensitivity of each of the thermal simulation parameters and the third response sensitivity of each of the parameter combinations, a sampling density of each of the thermal simulation parameters within a high-sensitivity interval is determined.

[0019] By adopting the above technical scheme, all thermal simulation parameters in the high-sensitivity interval are randomly combined to determine the influence of different sampling densities on the prediction results of the BP neural network model when all thermal simulation parameters contained in a single parameter combination are sampled according to the corresponding sampling density. If the influence significance increases with the increase of the sampling density, it can be considered that the corresponding parameter combination synergistically affects the prediction result, so that the sampling density of the thermal simulation parameters contained in the parameter combination can be defined by the sampling density, so as to achieve accurate determination of the sampling densities corresponding to different thermal simulation parameters in different data intervals.

[0020] In the second aspect, the present application provides an electric vehicle connector temperature rise prediction system based on LHS optimized GA-BP algorithm, comprising: A prediction model optimization module is used to obtain optimization instructions, construct an optimization data set formed by thermal simulation parameter values ​​of electric vehicle connectors based on a preset sampling density and a preset LHS sampling algorithm; and optimize a pre-constructed BP neural network model using the optimization data set and a preset GA algorithm; The temperature rise data prediction module is used to take the thermal simulation parameter values ​​contained in the prediction instruction as input parameters and input them into the latest optimized BP neural network model whenever a prediction instruction is received, and output the prediction result with temperature data through the BP neural network model; A sampling density update module is used to determine the first response sensitivity and sampling density of each data interval based on the response output result of the BP neural network model to the data set corresponding to each data interval during the optimization process; wherein the data interval refers to a number of sampling intervals formed by the LHS sampling algorithm according to the thermal simulation parameter value; the first response sensitivity is used to characterize the significance of the influence of the input variable of the corresponding data interval on the output result of the BP neural network model, and the more significant the influence, the higher the corresponding response sensitivity and the greater the sampling density; The sampling density updating module is further used to define a multi-order sampling density for each thermal simulation parameter in the high-sensitivity interval according to the first response sensitivity of each data interval, and determine the second response sensitivity and sampling density of each thermal simulation parameter in the high-sensitivity interval after sampling the thermal simulation parameters in the high-sensitivity interval step by step according to the corresponding multi-order sampling density; wherein the high-sensitivity interval refers to a data interval in which the first response sensitivity is higher than the preset sensitivity; the second response sensitivity is used to characterize the significance of the influence of the sampling density of the thermal simulation parameters in the high-sensitivity interval on the output result of the BP neural network model; The sampling density updating module is further used to update the preset sampling density of the non-high-sensitivity interval and the sampling density corresponding to each thermal simulation parameter of the high-sensitivity interval using the determined sampling density.

[0021] In a third aspect, the present application provides an electric vehicle connector temperature rise prediction device based on the LHS optimized GA-BP algorithm, comprising a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute any method described in the first aspect.

[0022] In a fourth aspect, the present application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and execute any one of the methods described in the first aspect.

[0023] In summary, this application includes the following beneficial technical effects: Traditional BP neural network is learned and trained through forward signal transmission and reverse error propagation. The model is prone to fall into the local optimal solution due to the randomness of initial weights and thresholds, resulting in limited stability and accuracy of prediction results. To overcome this limitation, LHS optimized GA-BP neural network is introduced. The weight and threshold configuration of BP neural network is optimized by genetic algorithm, and the LHS method is used to improve the initial diversity and global coverage of the population. The LHS method generates an initial population covering the whole world by uniformly partitioning and randomly sampling in the parameter space, thus avoiding the trap of local optimality in early iterations. The optimized GA-BP neural network model uses the selection, crossover and mutation process of genetic algorithm to further iterate, gradually improving the search ability and convergence effect of the network. Compared with traditional methods, LHS optimized GA-BP neural network greatly improves the global search ability and prediction accuracy of the model, showing excellent application value in the temperature management of electric vehicle connectors. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0025] Figure 1 It is a flow chart of a method for predicting temperature rise of an electric vehicle connector based on an LHS optimized GA-BP algorithm disclosed in an embodiment of the present application.

[0026] Figure 2 It is a schematic diagram used to illustrate the contact points of the connecting terminals in an embodiment of the present application.

[0027] Figure 3 It is a schematic diagram of the optimization logic of the GA-BP algorithm based on LHS optimization disclosed in the embodiment of the present application.

[0028] Figure 4 It is a structural block diagram of an electric vehicle connector temperature rise prediction system based on the LHS optimized GA-BP algorithm in an embodiment of the present application.

[0029] Description of reference numerals: 201, prediction model optimization module; 202, sampling density update module; 203, temperature rise data prediction module. DETAILED DESCRIPTION

[0030] The embodiment of the present application discloses a temperature rise prediction method for an electric vehicle connector based on an LHS optimized GA-BP algorithm (hereinafter referred to as the temperature rise prediction method). The temperature rise prediction method is used to train a BP neural network model optimized by a GA algorithm based on sample data obtained by LHS uniform sampling, and finally the temperature rise data of an electric vehicle connector is predicted by the BP neural network model after the aforementioned training optimization, thereby improving prediction efficiency and prediction accuracy. The execution subject of the temperature rise prediction method is a temperature rise prediction system for an electric vehicle connector based on an LHS optimized GA-BP algorithm (hereinafter referred to as the temperature rise prediction system). Figure 1-3 The specific implementation steps of the temperature rise prediction method of the temperature rise prediction system are specifically explained.

[0031] S101, obtaining optimization instructions, constructing an optimization data set formed by thermal simulation parameter values ​​of an electric vehicle connector based on a preset sampling density and a preset LHS sampling algorithm; and optimizing a pre-constructed BP neural network model using the optimization data set and a preset GA algorithm.

[0032] S102, whenever a prediction instruction is received, the thermal simulation parameter value contained in the prediction instruction is used as an input parameter and input into the latest optimized BP neural network model, and the prediction result with temperature data is output through the BP neural network model.

[0033] In implementation, the optimization instruction can be triggered manually or automatically by the temperature rise prediction system according to preset conditions. When the optimization instruction is triggered (i.e., when the temperature rise system obtains the optimization instruction), the temperature rise prediction system will use the predicted LHS sampling algorithm and perform sampling according to the preset sampling density. The sampling object is the specific value of the thermal simulation parameters of the electric vehicle connector. In the embodiment of the present application, the thermal simulation parameters at least include the charging time, the resistance value of the connection area of ​​the charging gun end, the resistance value of the connection area of ​​the charging seat end, the contact resistance value of the terminal matching area, the ambient temperature, and the temperature value of the connection terminal contact point. Among them, the connection terminal contact point can be multiple (such as Figure 2 ), the temperature value of the corresponding connection terminal contact point is the temperature value at each connection terminal contact point.

[0034] The temperature rise prediction system will generate an optimized data set for the thermal simulation parameter values ​​obtained by the above sampling, where the optimized data set contains several data groups, each data group includes input parameter values ​​and output parameter values, where the input parameter values ​​are charging time, resistance value of the connection area at the charging gun end, resistance value of the connection area at the charging base end, contact resistance value of the terminal mating area, and ambient temperature; the output parameter value is the temperature value of the contact point of the connection terminal.

[0035] Accordingly, refer to Figure 3, the BP neural network model pre-constructed in this application specifically includes an input layer, a hidden layer and an output layer, wherein the input layer includes 5 neurons, respectively for the above 5 input parameter values ​​(i.e., charging time, resistance value of the connection area at the charging gun end, resistance value of the connection area at the charging seat end, contact resistance value of the terminal matching area, and ambient temperature); the hidden layer is configured with 10 neurons, and an S-type activation function tansig is used. The output layer is configured with a number of neurons, and the number of neurons is consistent with the number of the above output parameter values ​​(i.e., the number of contact points of the connection terminals) and corresponds one to one. In an embodiment of the present application, the output layer is configured with 7 neurons, corresponding to the temperature values ​​of the 7 different contact points of the connection terminals of the electric vehicle charging connector, and a linear transfer function purelin is used. Among them, the output signal of the hidden layer is: Y k =g(∑V ik × i -b k );where Y k represents the output of the kth node in the hidden layer, V ik is the weight from the i-th node in the input layer to the k-th node in the hidden layer, z i is the input value of the i-th node in the input layer, b k represents the bias of the kth node in the hidden layer. The calculation formula of the output layer is: k =F(∑W jk ×H j -b k );where H j is the jth output signal, W jk is the weight from node j to node k, b k is the threshold of node k, O k The output signal of node k. It should be noted that the BP neural network model pre-constructed in the present application is a BP neural network model that has been preliminarily trained in advance, and the error converges to a preset threshold or no longer significantly decreases, and the prediction ability meets the standard.

[0036] Based on the pre-built BP neural network model, the GA algorithm is operated, and the initial population size is set to 100, the fitness value of each individual is calculated, and the error is used to determine whether the termination condition is met. Among them, the fitness function is defined as: In the formula, G j is the fitness value of the jth individual, u is the predicted output of the BP network, and v is the expected output.

[0037] In order to improve the diversity of the population, the mutation operation is introduced into the GA-BP algorithm to randomly adjust the gene values ​​of weights and biases. The crossover operation uses the gene fragment exchange method to increase the genetic diversity of individuals with higher fitness with a crossover probability of 0.6 and a mutation probability of 0.02 to ensure that it converges to the optimal solution in the end. The individuals with the highest fitness value are selected as offspring, and the roulette method is used to screen individuals with excellent fitness from the population, and genetic evolution is carried out in combination with the characteristics of both parents. Through mutation and crossover, the fitness of individuals is optimized generation by generation to enhance the global search ability; the weights and biases obtained by GA optimization are used as the initial parameters of the BP neural network, and the BP network is retrained until the error between the model output value and the training sample meets the preset target. At this point, the LHS optimized GA-BP neural network model can be used for prediction. Among them, when the roulette selection method is used to select individuals with good fitness from the population, the corresponding probability calculation formula is: In the formula, s j is the inverse value of the fitness of the jth individual, c is a constant, m is the total number of individuals, p j is the selection probability. When performing the crossover operation, a real number coded crossover method is used to generate new individuals. For the xth and yth individuals, the new individuals after the crossover are calculated as follows: z ′ x =z x (1-η)+z y η z ′ y =z y (1-η)+z x η Where η is a random number in the interval [0,1]. After the mutation operation and the generation of a new population, the individual gene values ​​can be adjusted according to the following rules: In the formula, h max and h min are the upper and lower bounds of the gene, β is a random factor, θ is a random number in the interval [0,1], and g is the current evolutionary generation. During the genetic evolution process, individuals with high fitness are retained and individuals with low fitness are replaced to update the population. The above steps are repeated until the fitness exceeds the predetermined threshold or the number of iterations reaches the set upper limit. The iteration is terminated and the current optimal weight threshold is recorded.

[0038] S103, based on the response output results of the BP neural network model to the data set corresponding to each data interval during the optimization process, determine the first response sensitivity and sampling density of each data interval; wherein, the data interval refers to a number of numerical intervals formed by dividing the thermal simulation parameters when sampling by the LHS sampling algorithm; the first response sensitivity is used to characterize the significance of the influence of the input variables of the corresponding data interval on the output results of the BP neural network model, and the more significant the influence, the higher the corresponding response sensitivity and the greater the sampling density.

[0039] S104, according to the first response sensitivity of each data interval, define a multi-order sampling density for each thermal simulation parameter in the high-sensitivity interval, and after sampling the thermal simulation parameters in the high-sensitivity interval step by step according to the corresponding multi-order sampling density, determine the second response sensitivity and sampling density of each thermal simulation parameter in the high-sensitivity interval; wherein, the high-sensitivity interval refers to a data interval whose first response sensitivity is higher than a preset sensitivity; the second response sensitivity is used to characterize the significance of the influence of the sampling density of the thermal simulation parameters in the high-sensitivity interval on the output result of the BP neural network model.

[0040] S105: using the determined sampling density to update the preset sampling density of the non-high-sensitivity interval and the sampling density corresponding to each thermal simulation parameter of the high-sensitivity interval.

[0041] In implementation, the temperature difference between the temperature rise data predicted and output by the BP neural network model and the preset temperature threshold can be used as the basis for determining the significance of the impact, and the first response sensitivity and the sampling density used in each data interval can be determined according to the pre-stored correspondence table between the temperature difference range, the first response sensitivity value and the sampling density. It is believed that: the smaller the temperature difference, the larger the corresponding first response sensitivity value, the more significant the corresponding impact, and the larger the sampling density to be configured, that is, the sampling density of the data interval where the predicted temperature rise data is closer to the preset temperature threshold is increased, thereby improving the prediction sensitivity and accuracy of the BP neural network model for the data in the data interval.

[0042] In addition, the difference between the temperature rise data output after the BP neural network model predicts the input parameters belonging to the same data interval (i.e., the temperature rise data difference, hereinafter referred to as the first difference) can also be used as the basis for determining the significance of the impact, and the first response sensitivity and the sampling density used in each data interval are determined according to the pre-stored correspondence table between the temperature rise data difference range, the first response sensitivity value and the sampling density. It is believed that: the larger the first difference, the larger the corresponding first response sensitivity value, the more significant the corresponding impact, and the larger the sampling density to be configured, that is, the variability of the temperature rise data predicted by the corresponding data interval is large, and the prediction sensitivity and accuracy of the BP neural network training for the data interval are enhanced by increasing the sampling density of such data intervals.

[0043] Accordingly, after determining the first response sensitivity and the corresponding sampling density of each data interval, for the data interval whose first response sensitivity is higher than the preset sensitivity (i.e., the high-sensitivity interval), the present application proposes to further refine its sampling density (hereinafter referred to as the initial sampling density), specifically to re-determine the sampling density of each thermal simulation parameter (herein refers to the input parameter value) in the high-sensitivity interval, and finally replace the preset sampling density according to the sampling density of the non-high-sensitivity interval and the sampling density determined for each thermal simulation parameter of each high-sensitivity interval, so as to use this as the sampling density for each input parameter when the optimization instruction is subsequently triggered.

[0044] Specifically, the sampling density of each thermal simulation parameter in the high-sensitivity range is determined as follows: According to the initial sampling density of the high-sensitivity interval, a multi-order sampling density ρ is defined for all thermal simulation parameters (i.e., input parameter values) in the high-sensitivity interval, and the initial sampling density is used as the lowest-order sampling density. Each order sampling density value is different, and the higher the order, the larger the sampling density value. Then, all the order sampling density values ​​included in the multi-order sampling density are used to sample the thermal simulation parameters step by step until all the thermal simulation parameters (i.e., input parameter values) are sampled according to all the order sampling density values. The input parameters of each data group contained in the corresponding optimized data set can be specifically expressed as: (charging time ρ a , the resistance value of the charging gun connection area ρ b , the resistance value of the charging dock connection area ρ c , Terminal mating area contact resistance value ρ d 、Ambient temperature ρ d ), wherein a, b, c, and d represent the sampling density of the corresponding input parameters in the data set. At the same sampling time, the order sampling density values ​​used for different thermal simulation parameters may be the same or different.

[0045] The temperature rise prediction system is used to determine the second response sensitivity and sampling density of the input parameter of charging time according to the charging time obtained by sampling different a when bcd is the same, and the change of the temperature rise data predicted and output by the BP neural network. The greater the difference (hereinafter referred to as the second difference) between the prediction results (i.e., temperature rise data) obtained after sampling the charging time with different sampling densities (i.e., a), the greater the impact significance. The second response sensitivity of the charging time and the corresponding sampling density can be determined from the aforementioned corresponding relationship according to the correspondence between the range of several pre-stored differences, the second response sensitivity and the sampling density, and according to the maximum second temperature difference. In the above manner, the sampling density of all input parameters in the high-sensitivity interval can be determined, that is, the sampling density of other input parameters except the target input parameter remains unchanged, and the sampling density of the target input parameter is switched, and the input parameter value obtained by the sampling density is input into the BP neural network to realize the prediction, and the second response sensitivity and sampling density are determined according to the pre-second difference; the target input parameter here refers to any input parameter. Finally, this operation is used to further identify the input parameters that have a significant impact on the prediction results from the high-sensitivity interval and enhance their sampling density, thereby helping to improve the prediction accuracy of the BP neural network model.

[0046] Optionally, determining the second response sensitivity and sampling density of each thermal simulation parameter in the high-sensitivity interval further includes: randomly combining all thermal simulation parameters in the high-sensitivity interval to form a parameter combination, sampling each of the parameter combinations in the high-sensitivity interval step by step according to the corresponding multi-order sampling density, and determining the third response sensitivity of each of the parameter combinations; wherein the third response sensitivity is used to characterize the significance of the influence of the sampling density of the parameter combination in the high-sensitivity interval on the output result of the BP neural network model; Based on the second response sensitivity of each of the thermal simulation parameters and the third response sensitivity of each of the parameter combinations, a sampling density of each of the thermal simulation parameters within a high-sensitivity interval is determined.

[0047] In the implementation, all thermal simulation parameters in the high-sensitivity range are randomly combined to form a parameter combination, and the number of input parameters contained in the parameter combination is more than one arbitrary value. Then, similarly, the sampling density of all input parameters except the input parameters contained in the parameter combination is controlled to remain unchanged, and the corresponding input parameters are sampled by switching the sampling density of the input parameters contained in the target parameter combination. Then, according to the input parameter values ​​obtained by sampling (such as (charging time ρ a , the resistance value of the charging gun connection area ρ b), and a and b can be the same or different), the BP neural network model is used to predict the output temperature rise data, and the difference in temperature rise data corresponding to different sampling densities (hereinafter referred to as the third difference) is used to evaluate the significance of the impact on the predicted temperature rise data. The temperature rise prediction system can pre-store different ranges of the third difference, and the third response sensitivity value corresponding to each range, and the sampling density corresponding to each third corresponding sensitivity value. Among them, the larger the third difference, the larger the corresponding third response sensitivity. When the third response sensitivity is higher than the preset sensitivity threshold, the current sampling density is used to define the sampling density of the input parameters contained in the corresponding parameter combination. This operation is used to find the parameter combination that has a significant impact on the temperature rise data from the high-sensitivity interval, and the prediction accuracy of the BP neural network model is optimized by adjusting its sampling density.

[0048] Optionally, the temperature rise prediction method further includes: Acquire charging information based on the onboard system of the electric vehicle connected in advance in real time, and determine the charging efficiency of the electric vehicle connector according to the charging information, and determine whether the change satisfies the preset trigger condition when the charging efficiency changes and the change is not a transient change based on the charging efficiency acquired in real time, and if so, trigger the optimization instruction; Record the BP neural network model after each optimization and the corresponding charging efficiency, and determine the effective prediction range of the BP neural network model.

[0049] S102 specifically includes the following sub-steps: S1021, whenever a prediction instruction is received, determining whether the prediction time is within the effective prediction range of the currently optimized BP neural network model according to the prediction time included in the prediction instruction; S1022, if yes, the thermal simulation parameter values ​​contained in the prediction instruction are used as input parameters and input into the latest optimized BP neural network model, and the prediction results with temperature data are output through the BP neural network model; if no, the prediction results with temperature data are output based on the preset proxy model prediction.

[0050] In implementation, the temperature rise prediction system is pre-communicated with the electric vehicle onboard system, and the electric vehicle onboard system can monitor and know the battery power of the electric vehicle in real time. Therefore, when the electric vehicle connector is connected to the electric vehicle for charging, the temperature rise prediction system can communicate with the corresponding electric vehicle onboard system in real time to know the charging efficiency of the electric vehicle connector. The charging efficiency can be considered as the charging amount per unit time. When the charging connector is used for a long time and becomes aged, or there are interruptions such as poor contact, its corresponding performance will decline, and the aforementioned conditions will also affect the temperature rise of the electric vehicle connector; therefore, this application uses the determination of whether the performance of the electric vehicle connector has changed by determining whether the charging efficiency has changed, and when the change occurs and the change meets the preset trigger condition, the optimization instruction is triggered to re-optimize the BP neural network model according to the thermal simulation parameter values ​​collected after the performance of the electric vehicle connector changes, so that the BP neural network model can adaptively adjust its prediction ability according to the performance change of the electric vehicle connector to ensure the prediction accuracy.

[0051] Specifically, the temperature rise prediction system will obtain the charging efficiency in real time. If the difference between the charging efficiency obtained at the target time and any of the charging efficiencies obtained in the specified time period before the target time is greater than the preset difference, it is considered that the charging efficiency has changed at the target time and the change meets the preset trigger condition, and if the charging efficiency corresponding to the target time appears more than the preset number of times in the specified time period after the target time, it is considered that the change in charging efficiency at the target time is not a short-term change. The target time refers to any moment.

[0052] Whenever the optimization instruction is triggered and the optimization is completed, the temperature rise prediction system is used to store the optimized BP neural network model and assign it the corresponding charging efficiency. The charging efficiency is reflected in a numerical range. For example, the charging efficiency t that has changed and meets the preset trigger conditions is taken as the midpoint of the range, and the numerical range is defined as [tm, t+m], where m is a specific value, and this range corresponds to the effective prediction range of the BP neural network model.

[0053] The temperature rise prediction system is used to determine whether the prediction time contained in the prediction instruction is within the effective prediction range of the current latest optimized BP neural network model each time a prediction instruction is received. The prediction time is specifically a time range and is used to represent the charging period. The corresponding prediction instruction is to apply for the prediction of the temperature rise data of the electric vehicle connector during the charging period, that is, the prediction time refers to a certain period in the future.

[0054] The specific determination scheme is as follows. Accordingly, S1021 specifically includes the following steps: S10211, obtaining a changing trend of charging efficiency over time, and predicting a model effective period corresponding to the charging efficiency corresponding to the currently optimized BP neural network model, where the model effective period refers to a time period for maintaining the corresponding charging efficiency; S10212, if the prediction time is within the effective period of the model, it is considered to be within the effective prediction range of the currently optimized BP neural network model; otherwise, it is considered not to be within the effective prediction range of the currently optimized BP neural network model.

[0055] In implementation, the temperature rise prediction system is used to obtain the changing trend of charging efficiency over time, that is, the changing speed of charging efficiency, in real time, and predict the corresponding effective period of the model according to the charging efficiency corresponding to the optimized BP neural network model. The effective period of the model is the time period in the future, and the corresponding charging efficiency is the charging efficiency of the current BP neural network. Then, it is determined whether the predicted time is within the effective period of the model. If so, it is considered that the predicted time is within the effective prediction range of the current optimized BP neural network model, and it can be used for prediction by the current optimized BP neural network model. Otherwise, it is not. Correspondingly, from the perspective of prediction accuracy, the temperature rise data predicted by the current optimized BP neural network model cannot be used as the final prediction result.

[0056] For prediction instructions whose prediction time is within the effective prediction range of the currently optimized BP neural network model, the temperature rise prediction system is used to input the thermal simulation parameter values ​​contained in the prediction instructions as input parameters into the currently optimized BP neural network, and then output the prediction results with temperature data through the BP neural network model.

[0057] For prediction instructions whose prediction time is not within the effective prediction range of the currently optimized BP neural network model, the prediction results with temperature data are output based on the preset proxy model prediction. The corresponding specific prediction scheme is as follows: The temperature rise prediction method also includes the following steps: Whenever the BP neural network model is optimized, the temperature deviation of the BP neural network model corresponding to the optimized model at different times is determined, and the pre-constructed proxy model is trained with the charging efficiency difference as the input parameter and the temperature deviation as the output parameter; if not in S1022, the prediction result with temperature data is output based on the preset proxy model prediction, including the following sub-steps: If not, predicting and outputting initial temperature data based on the thermal simulation parameter values ​​included in the prediction instruction through the current BP neural network model; Determine whether the prediction time has an intersection with the model effective period corresponding to the current BP neural network model; If there is no intersection, determining a first charging efficiency difference between the first charging efficiency and the charging efficiency corresponding to the current BP neural network model according to the first charging efficiency predicted corresponding to the predicted time, outputting a first temperature deviation based on the first charging efficiency difference through a proxy model, fusing the initial temperature data with the first temperature deviation to obtain temperature data, and outputting a prediction result with the temperature data; If there is an intersection, the time period corresponding to the intersection in the forecast time is taken as the effective forecast period, and the time period of the non-effective forecast period is taken as the deviation forecast period; The initial temperature data is used as the temperature data of the effective prediction period; the second charging efficiency predicted corresponding to the deviation prediction period is determined, and the second charging efficiency difference between the second charging efficiency and the charging efficiency corresponding to the current BP neural network model is determined, the second temperature deviation is output based on the second charging efficiency difference through the proxy model, and the initial temperature data and the second temperature deviation are merged to obtain the temperature data corresponding to the deviation prediction period; the prediction result with the effective prediction period, the deviation prediction period and its corresponding temperature data is output. In implementation, the temperature rise prediction system is also used to record the optimization time after each optimization to obtain a new BP neural network model, and then input the same input parameter value into the BP neural network model corresponding to different optimization times, calculate the difference in temperature rise data output by any two BP neural network models (i.e., temperature deviation), and the difference in charging efficiency corresponding to the two BP neural network models (i.e., charging efficiency difference), and use the charging efficiency difference as the input parameter and the temperature deviation as the output parameter to train the pre-constructed proxy model, that is, the proxy model is used to predict the output temperature deviation according to the charging efficiency difference, that is, predict the deviation of the prediction results of the BP neural network model corresponding to different optimization times.

[0058] Correspondingly, when the proxy model is needed to predict the temperature rise data corresponding to the current prediction instruction, the temperature rise prediction system will first predict the temperature rise data (i.e., initial temperature data) through the current BP neural network model according to the thermal simulation parameters contained in the prediction instruction. Then determine whether there is an intersection between the prediction time in the prediction instruction and the model effective period corresponding to the current BP neural network model. If there is no intersection, then according to the above-mentioned: the trend of charging efficiency over time, predict the charging efficiency corresponding to the prediction time (i.e., the first charging efficiency), and then input the difference between the first charging efficiency and the charging efficiency corresponding to the current BP neural network model (i.e., the first charging efficiency difference) into the proxy model, output the first temperature deviation through the proxy model, sum the initial temperature data and the first temperature deviation to obtain the temperature data, output the prediction result with the temperature data, and complete the prediction.

[0059] If there is an intersection, the part of the prediction time with the intersection is used as the effective prediction period, and the time period of the prediction time that is not the effective prediction period is used as the deviation prediction period; the initial temperature data is the temperature rise data corresponding to the effective prediction period, and for the deviation prediction period, according to the above-mentioned trend of charging efficiency changes over time, the charging efficiency corresponding to the deviation prediction period (i.e., the second charging efficiency) is predicted, and then the difference between the second charging efficiency and the charging efficiency corresponding to the current BP neural network model (i.e., the second charging efficiency difference) is input into the proxy model, and the second temperature deviation is output through the proxy model. The initial temperature data and the second temperature deviation are summed to obtain the temperature data, which is the temperature data corresponding to the deviation prediction period. Finally, the prediction result with the effective prediction period, the deviation prediction period and its corresponding Wendy data is output.

[0060] The present application also discloses a temperature rise prediction system for electric vehicle connectors based on the LHS optimized GA-BP algorithm. Figure 4 , the temperature rise prediction of electric vehicle connectors based on LHS optimized GA-BP algorithm includes: The prediction model optimization module 201 is used to obtain optimization instructions, construct an optimization data set formed by thermal simulation parameter values ​​of the electric vehicle connector based on a preset sampling density and a preset LHS sampling algorithm; and optimize the pre-constructed BP neural network model using the optimization data set and a preset GA algorithm; The temperature rise data prediction module 202 is used to take the thermal simulation parameter value contained in the prediction instruction as an input parameter and input it into the latest optimized BP neural network model whenever a prediction instruction is received, and output the prediction result with temperature data through the BP neural network model; The sampling density updating module 203 is used to determine the first response sensitivity and sampling density of each data interval based on the response output result of the BP neural network model to the data set corresponding to each data interval during the optimization process; wherein the data interval refers to a number of sampling intervals formed by the LHS sampling algorithm according to the thermal simulation parameter value; the first response sensitivity is used to characterize the significance of the impact of the input variable of the corresponding data interval on the output result of the BP neural network model, and the more significant the impact, the higher the corresponding response sensitivity and the greater the sampling density; The sampling density updating module 203 is further used to define a multi-order sampling density for each thermal simulation parameter in the high-sensitivity interval according to the first response sensitivity of each data interval, and determine the second response sensitivity and sampling density of each thermal simulation parameter in the high-sensitivity interval after sampling the thermal simulation parameters in the high-sensitivity interval step by step according to the corresponding multi-order sampling density; wherein the high-sensitivity interval refers to a data interval in which the first response sensitivity is higher than the preset sensitivity; the second response sensitivity is used to characterize the significance of the influence of the sampling density of the thermal simulation parameters in the high-sensitivity interval on the output result of the BP neural network model; The sampling density updating module 203 is further configured to update the preset sampling density of the non-high-sensitivity interval and the sampling density corresponding to each thermal simulation parameter of the high-sensitivity interval using the determined sampling density.

[0061] Optionally, it also includes an optimization trigger module, which is used to obtain charging information in real time based on the electric vehicle's on-board system that is pre-communicated and connected, and determine the charging efficiency of the electric vehicle connector based on the charging information. When the charging efficiency meets the preset trigger conditions, the optimization instruction is triggered; it is also used to record the BP neural network model after each optimization, as well as the corresponding charging efficiency, and determine the effective prediction range of the BP neural network model.

[0062] The temperature rise data prediction module 202 is also used to determine whether the prediction time is within the effective prediction range of the currently optimized BP neural network model based on the prediction time contained in the prediction instruction whenever a prediction instruction is received; if so, the thermal simulation parameter value contained in the prediction instruction is used as an input parameter and input into the latest optimized BP neural network model, and the prediction result with temperature data is output through the BP neural network model; if not, the prediction result with temperature data is output based on the preset proxy model prediction.

[0063] Optionally, the temperature rise data prediction module 202 is also used to obtain the changing trend of the charging efficiency over time, and predict the model effective period corresponding to the charging efficiency corresponding to the currently optimized BP neural network model. The model effective period refers to the time period for maintaining the corresponding charging efficiency; if the predicted time is within the model effective period, it is considered to be within the effective prediction range of the currently optimized BP neural network model, otherwise it is considered not to be within the effective prediction range of the currently optimized BP neural network model.

[0064] Optionally, it also includes a proxy model training module, which is used to determine the temperature deviation of the optimized BP neural network model corresponding to different times every time the BP neural network model is optimized, and to train the pre-constructed proxy model with the charging efficiency difference as the input parameter and the temperature deviation as the output parameter.

[0065] The temperature rise data prediction module 202 is also used to, if not, use the thermal simulation parameter value contained in the prediction instruction as an input parameter and input it into the latest optimized BP neural network model, and output the initial temperature data through the BP neural network model; it is also used to predict the corresponding charging efficiency according to the prediction time, and determine the charging efficiency difference between the charging efficiency corresponding to the prediction time and the charging efficiency corresponding to the current BP neural network model, and output the temperature deviation based on the charging efficiency difference through the proxy model; it is also used to fuse the initial temperature data with the temperature deviation to obtain temperature data, and output the prediction result with temperature data.

[0066] The temperature rise data prediction module 202 is also used to predict and output initial temperature data based on the thermal simulation parameter values ​​contained in the prediction instruction through the current BP neural network model; determine whether there is an intersection between the prediction time and the model effective period corresponding to the current BP neural network model; if there is no intersection, determine the first charging efficiency difference between the first charging efficiency and the charging efficiency corresponding to the current BP neural network model according to the first charging efficiency predicted corresponding to the prediction time, output the first temperature deviation based on the first charging efficiency difference through the proxy model, fuse the initial temperature data with the first temperature deviation to obtain temperature data, and output the prediction result with temperature data; if there is Intersection, then the time period corresponding to the intersection part in the prediction time is used as the effective prediction period, and the time period of the non-effective prediction period is used as the deviation prediction period; it is also used to use the initial temperature data as the temperature data of the effective prediction period; determine the second charging efficiency corresponding to the deviation prediction period, and the second charging efficiency difference between the second charging efficiency and the charging efficiency corresponding to the current BP neural network model, output the second temperature deviation based on the second charging efficiency difference through the proxy model, merge the initial temperature data with the second temperature deviation to obtain the temperature data corresponding to the deviation prediction period; output the prediction result with the effective prediction period, the deviation prediction period and its corresponding Wendy data.

[0067] Optionally, the optimization trigger module is also used to determine whether the change meets a preset trigger condition based on the charging efficiency acquired in real time when the charging efficiency changes and the change is not a transient change. If so, the optimization instruction is triggered.

[0068] Optionally, the sampling density update module 203 is also used to randomly combine all thermal simulation parameters in the high-sensitivity interval to form a parameter combination, and to sample each parameter combination in the high-sensitivity interval step by step according to the corresponding multi-order sampling density to determine the third response sensitivity of each parameter combination; wherein the third response sensitivity is used to characterize the significance of the influence of the sampling density of the parameter combination in the high-sensitivity interval on the output result of the BP neural network model; and is also used to determine the sampling density of each thermal simulation parameter in the high-sensitivity interval based on the second response sensitivity of each thermal simulation parameter and the third response sensitivity of each parameter combination.

[0069] An embodiment of the present application also discloses a temperature rise prediction device for an electric vehicle connector based on an LHS optimized GA-BP algorithm. The temperature rise prediction device for an electric vehicle connector based on an LHS optimized GA-BP algorithm includes a memory and a processor. The memory stores a computer program that can be loaded by the processor and executes the above-mentioned method for predicting the temperature rise of an electric vehicle connector based on the LHS optimized GA-BP algorithm.

[0070] An embodiment of the present application also discloses a computer-readable storage medium, which stores a computer program that can be loaded by a processor and executes the above-mentioned electric vehicle connector temperature rise prediction method based on the LHS optimized GA-BP algorithm. The computer-readable storage medium includes, for example: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0071] It should be noted that, in this document, relational terms such as first and second, etc. are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations.

[0072] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit the protection scope of the application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on these embodiments, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

Claims

1. A method for predicting temperature rise of electric vehicle connectors based on LHS optimized GA-BP algorithm, characterized in that: include: Obtain optimization instructions, and construct an optimization data set formed by thermal simulation parameter values ​​of the electric vehicle connector based on a preset sampling density and a preset LHS sampling algorithm; Optimizing the pre-built BP neural network model using the optimized data set and a preset GA algorithm; Whenever a prediction instruction is received, the thermal simulation parameter value contained in the prediction instruction is used as an input parameter and input into the latest optimized BP neural network model, and the prediction result with temperature data is output through the BP neural network model; Based on the response output results of the BP neural network model to the data set corresponding to each data interval during the optimization process, the first response sensitivity and sampling density of each data interval are determined; wherein the data interval refers to a number of numerical intervals formed by dividing the thermal simulation parameters when sampling by the LHS sampling algorithm; the first response sensitivity is used to characterize the significance of the influence of the input variables of the corresponding data interval on the output results of the BP neural network model, and the more significant the influence, the higher the corresponding response sensitivity and the greater the sampling density; According to the first response sensitivity of each data interval, a multi-order sampling density is defined for each thermal simulation parameter in the high-sensitivity interval, and after sampling the thermal simulation parameters in the high-sensitivity interval step by step according to the corresponding multi-order sampling density, a second response sensitivity and sampling density of each thermal simulation parameter in the high-sensitivity interval are determined; wherein the high-sensitivity interval refers to a data interval in which the first response sensitivity is higher than the preset sensitivity; the second response sensitivity is used to characterize the significance of the influence of the sampling density of the thermal simulation parameters in the high-sensitivity interval on the output result of the BP neural network model; The determined sampling density is used to update the preset sampling density of the non-high-sensitivity interval and the sampling density corresponding to each thermal simulation parameter of the high-sensitivity interval.

2. The method for predicting temperature rise of electric vehicle connector based on LHS optimized GA-BP algorithm according to claim 1 is characterized in that: The method further comprises: The charging information is acquired in real time based on the onboard system of the electric vehicle connected in advance, and the charging efficiency of the electric vehicle connector is determined according to the charging information. When the charging efficiency meets the preset trigger condition, the optimization instruction is triggered; Recording the BP neural network model after each optimization and the corresponding charging efficiency, and determining the effective prediction range of the BP neural network model; Whenever a prediction instruction is received, the thermal simulation parameter value contained in the prediction instruction is used as an input parameter and input into the latest optimized BP neural network model, and the prediction result with temperature data is outputted by the BP neural network model, including: Whenever a prediction instruction is received, determining whether the prediction time is within the effective prediction range of the currently optimized BP neural network model according to the prediction time included in the prediction instruction; If so, the thermal simulation parameter values ​​contained in the prediction instruction are used as input parameters and input into the latest optimized BP neural network model, and the prediction results with temperature data are output through the BP neural network model; if not, the prediction results with temperature data are output based on the preset proxy model prediction.

3. The temperature rise prediction method of electric vehicle connector based on LHS optimized GA-BP algorithm according to claim 2 is characterized in that: Whenever a prediction instruction is received, determining whether the prediction time is within the effective prediction range of the currently optimized BP neural network model according to the prediction time included in the prediction instruction includes: Obtain the changing trend of charging efficiency over time, and predict the effective period of the model corresponding to the charging efficiency corresponding to the currently optimized BP neural network model, where the effective period of the model refers to the time period for maintaining the corresponding charging efficiency; If the prediction time is within the effective period of the model, it is considered to be within the effective prediction range of the currently optimized BP neural network model; otherwise, it is considered not to be within the effective prediction range of the currently optimized BP neural network model.

4. The temperature rise prediction method for electric vehicle connector based on LHS optimized GA-BP algorithm according to claim 3 is characterized in that: The method further comprises: Whenever the BP neural network model is optimized, the temperature deviation of the optimized BP neural network model corresponding to different times is determined, and the pre-constructed proxy model is trained with the charging efficiency difference as an input parameter and the temperature deviation as an output parameter; If not, the prediction result with temperature data is output based on the preset proxy model prediction, including: If not, the thermal simulation parameter values ​​contained in the prediction instruction are used as input parameters and input into the latest optimized BP neural network model, and the initial temperature data is output through the BP neural network model; Predicting a charging efficiency corresponding to the predicted time, determining a charging efficiency difference between the charging efficiency corresponding to the predicted time and the charging efficiency corresponding to the current BP neural network model, and outputting a temperature deviation based on the charging efficiency difference through a proxy model; The initial temperature data and the temperature deviation are fused to obtain temperature data, and a prediction result including the temperature data is output.

5. The method for predicting temperature rise of electric vehicle connector based on LHS optimized GA-BP algorithm according to claim 3 is characterized in that: The thermal simulation parameter values ​​included in the prediction instruction include at least the charging duration; the prediction time is the charging period corresponding to the combination of the charging start time and the charging duration; If not, the thermal simulation parameter value contained in the prediction instruction is used as an input parameter and input into the latest optimized BP neural network model, and the initial temperature data is output through the BP neural network model; the charging efficiency corresponding to the prediction time is predicted according to the prediction time, and the charging efficiency difference between the charging efficiency corresponding to the prediction time and the charging efficiency corresponding to the current BP neural network model is determined, and the temperature deviation is output based on the charging efficiency difference through the proxy model, the initial temperature data and the temperature deviation are merged to obtain temperature data, and the prediction result with the temperature data is output, including: If not, predicting and outputting initial temperature data based on the thermal simulation parameter values ​​included in the prediction instruction through the current BP neural network model; Determine whether the prediction time has an intersection with the model effective period corresponding to the current BP neural network model; If there is no intersection, determining a first charging efficiency difference between the first charging efficiency and the charging efficiency corresponding to the current BP neural network model according to the first charging efficiency predicted corresponding to the predicted time, outputting a first temperature deviation based on the first charging efficiency difference through a proxy model, fusing the initial temperature data with the first temperature deviation to obtain temperature data, and outputting a prediction result with the temperature data; If there is an intersection, the time period corresponding to the intersection in the forecast time is taken as the effective forecast period, and the time period of the non-effective forecast period is taken as the deviation forecast period; The initial temperature data is used as the temperature data of the effective prediction period; the second charging efficiency predicted corresponding to the deviation prediction period and the second charging efficiency difference between the second charging efficiency and the charging efficiency corresponding to the current BP neural network model are determined, and the second temperature deviation is output based on the second charging efficiency difference through the proxy model, and the initial temperature data and the second temperature deviation are merged to obtain the temperature data corresponding to the deviation prediction period; the prediction result with the effective prediction period, the deviation prediction period and its corresponding Wendy data is output.

6. The method for predicting temperature rise of electric vehicle connector based on LHS optimized GA-BP algorithm according to claim 2 is characterized in that: When the charging efficiency meets the preset trigger condition, triggering the optimization instruction includes: Based on the charging efficiency acquired in real time, when the charging efficiency changes and the change is not a transient change, it is determined whether the change meets a preset trigger condition, and if so, an optimization instruction is triggered.

7. The method for predicting temperature rise of electric vehicle connector based on LHS optimized GA-BP algorithm according to claim 1 is characterized in that: The determining of the second response sensitivity and sampling density of each thermal simulation parameter within the high-sensitivity interval includes: All thermal simulation parameters in the high-sensitivity interval are randomly combined to form a parameter combination, and each of the parameter combinations in the high-sensitivity interval is sampled step by step according to the corresponding multi-order sampling density to determine the third response sensitivity of each of the parameter combinations; wherein the third response sensitivity is used to characterize the significance of the influence of the sampling density of the parameter combination in the high-sensitivity interval on the output result of the BP neural network model; Based on the second response sensitivity of each of the thermal simulation parameters and the third response sensitivity of each of the parameter combinations, a sampling density of each of the thermal simulation parameters within a high-sensitivity interval is determined.

8. An electric vehicle connector temperature rise prediction system based on LHS optimized GA-BP algorithm, characterized in that: include: A prediction model optimization module (201) is used to obtain optimization instructions, and to construct an optimization data set formed by thermal simulation parameter values ​​of the electric vehicle connector based on a preset sampling density and a preset LHS sampling algorithm; and to optimize the pre-constructed BP neural network model using the optimization data set and a preset GA algorithm; A temperature rise data prediction module (202) is used to, whenever a prediction instruction is received, use the thermal simulation parameter value contained in the prediction instruction as an input parameter and input it into the latest optimized BP neural network model, and output a prediction result with temperature data through the BP neural network model; A sampling density updating module (203) is used to determine the first response sensitivity and sampling density of each data interval based on the response output result of the BP neural network model to the data set corresponding to each data interval during the optimization process; wherein the data interval refers to a plurality of sampling intervals formed by the LHS sampling algorithm according to the thermal simulation parameter value; the first response sensitivity is used to characterize the significance of the influence of the input variable of the corresponding data interval on the output result of the BP neural network model, and the more significant the influence, the higher the corresponding response sensitivity and the greater the sampling density; The sampling density updating module (203) is further used to define a multi-order sampling density for each thermal simulation parameter in the high-sensitivity interval according to the first response sensitivity of each data interval, and to determine the second response sensitivity and sampling density of each thermal simulation parameter in the high-sensitivity interval after sampling the thermal simulation parameters in the high-sensitivity interval step by step according to the corresponding multi-order sampling density; wherein the high-sensitivity interval refers to a data interval in which the first response sensitivity is higher than the preset sensitivity; and the second response sensitivity is used to characterize the significance of the influence of the sampling density of the thermal simulation parameters in the high-sensitivity interval on the output result of the BP neural network model; The sampling density updating module (203) is also used to update the preset sampling density of the non-high-sensitivity interval and the sampling density corresponding to each thermal simulation parameter of the high-sensitivity interval using the determined sampling density.

9. An electric vehicle connector temperature rise prediction device based on LHS optimized GA-BP algorithm, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute any one of the methods according to claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor and execute any one of the methods as claimed in claims 1 to 7.

Images