Knowledge data collaboratively-driven electric vehicle driving range hierarchical prediction method

Abstract

The invention is suitable for the technical field of electric vehicles, and provides a hierarchical prediction method for the driving range of an electric vehicle cooperatively driven by knowledge data. According to the method, the influence of energy consumption on the driving mileage of the electric vehicle is considered, and the influence of available energy on the driving mileage is also considered, so that the prediction precision of the driving mileage of the electric vehicle is improved; the electric vehicle driving range prediction problem is converted into the electric vehicle energy consumption rate and actual energy coefficient prediction problem by adopting the hierarchical prediction model, so that the flexibility and expandability of the model are improved, the development time and cost are reduced, and the accuracy of electric vehicle driving range prediction is greatly improved. In addition, through a model cooperatively driven by data and a mechanism, the limitation of a physical model is overcome, the interpretability is enhanced, the dependence on large-scale data is reduced, the problem of model overfitting is effectively avoided, the trust degree of a user is enhanced, and popularization of the electric vehicle is facilitated.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric vehicles, and in particular relates to a knowledge data collaboratively driven stratified prediction method for electric vehicle driving range. Background Art

[0002] In recent years, the rapid development of electric vehicles has benefited from the promotion of national policies, the growth of market demand and the rapid advancement of technology. The electrification of transportation is seen as a key way to improve energy efficiency and reduce carbon emissions. However, in the process of promoting electric vehicles, driving range anxiety has become a major concern for consumers. They are worried that insufficient battery life will lead to power exhaustion during long-distance driving, and it will be difficult to find charging facilities, affecting the completion of the journey.

[0003] Driving range refers to the maximum distance an electric vehicle can travel continuously under different road or environmental conditions when the battery is fully charged. It is an important indicator for measuring electric vehicle battery performance and vehicle energy efficiency, usually measured in kilometers (km) or miles (miles). However, in cold climates and complex driving conditions, the battery performance and energy consumption rate of electric vehicles will fluctuate significantly, making accurate prediction of electric vehicle driving range a major challenge facing the current industry. Traditional electric vehicle driving range prediction methods have the following main problems: First, traditional electric vehicle range prediction methods usually only consider the impact of electric vehicle energy consumption, and often ignore the impact of available energy on electric vehicle range. For example, battery degradation and low temperatures in winter lead to reduced available battery energy. These factors have a certain impact on the range. This neglect leads to low prediction accuracy of traditional models.

[0004] Secondly, traditional electric vehicle range prediction models usually adopt a single-layer structure, which concentrates all factors on one level for processing. This design not only reduces the efficiency of the model, but also limits the flexibility of the model, making it difficult to expand modularly as demand changes.

[0005] In addition, data-driven models are highly data-dependent, lack interpretability and extrapolation capabilities, and perform poorly in data-scarce or new scenarios. Although models based on mechanistic knowledge have advantages in interpretability and physical consistency, their modeling is complex, parameters are difficult to estimate, and they have difficulty dealing with complex nonlinear relationships, and their generalization capabilities are also poor.

[0006] In view of the above problems, the present invention proposes a knowledge data collaboratively driven stratified prediction method for electric vehicle driving range. Summary of the invention

[0007] The purpose of the present invention is to provide a knowledge data collaboratively driven electric vehicle driving range hierarchical prediction method, aiming to solve the problems raised in the above background technology.

[0008] To achieve the above object, the present invention provides the following technical solutions: A knowledge data collaboratively driven stratified prediction method for electric vehicle driving range includes the following steps: Step S1, collecting input features, including historical driving data, vehicle speed, battery temperature and battery status; Step S2, converting the problem of predicting the driving range of the electric vehicle into the problem of predicting the energy consumption rate and actual energy coefficient of the electric vehicle; Step S3: combining the prediction results of the energy consumption rate and the actual energy coefficient to generate a predicted value of the driving range of the electric vehicle.

[0009] Furthermore, the specific process of step S2 is as follows: Step 21: Use the MLP model to predict the energy consumption rate at the beginning of the trip, and use the LSTM model to predict the energy consumption rate during the driving phase. Step 22: First, use the data of each charging process to obtain the real actual energy coefficient label, and use the Monte Carlo simulation method to improve the credibility of the label; then use the temperature, constant current charging segment data and mileage data as input features to perform fast and slow charging identification, and predict the fast charging actual energy coefficient and slow charging actual energy coefficient respectively; finally, use the LSTM model to learn the time series changes of the actual energy coefficient during vehicle driving.

[0010] Furthermore, the specific method of step 21 is: first, use each discharge process data to obtain the real energy consumption rate label; then use temperature, driving information and start and end SOC as input features to calculate the energy consumption of each trip segment; then use the structure of switching between the MLP model and the LSTM model for prediction; at the beginning of the trip, due to insufficient data, the MLP model is used to call historical data for prediction. When the data volume reaches a preset threshold, the MLP model autonomously switches to the LSTM model to learn the time series changes of the collected driving data; finally, hyperparameter tuning and K-fold cross validation are used to improve the prediction performance of the model.

[0011] Furthermore, the specific method of step 22 is as follows: ; in, is the input gate; For the Gate of Oblivion; is the output gate; is the time step t The candidate memory cell states; is the time step t The unit status; is the time step t The hidden state of is the activation function; is the input weight matrix corresponding to the input gate; is the input weight matrix corresponding to the forget gate; Is the input weight matrix corresponding to the output gate; is the input weight matrix corresponding to the candidate memory unit state; is the current time step t The input vector of is the recursive weight matrix of the input gate; is the recursive weight matrix of the forget gate; is the recursive weight matrix of the output gate; is the recursive weight matrix of the candidate memory unit; is the time step t- The hidden state of 1; is the bias vector of the input gate; is the bias vector of the forget gate; is the bias vector of the output gate; is the bias vector of the candidate memory cell state; is the time step t The forget gate activation value of is the time step t- 1's unit status; is the time step t The input gate activation value of is the time step t The output gate activation value of .

[0012] Furthermore, the specific method of step S3 is as follows: Combining the prediction results of the energy consumption rate and the actual energy coefficient, the predicted value of the driving range of the electric vehicle is output through the MLP model, and the MLP model includes an input layer, a hidden layer and an output layer; The input layer receives input features. Assume that the input vector is , X is the entire input feature vector, including all input features; x1,x2,…,x n For each element in the input feature vector, that is, each feature or variable; n is the number of features; the input layer passes the feature vector to the first hidden layer; The neurons in the hidden layer receive the input features of the first input layer and perform linear transformation through weighted summation: ; in, Indicates l Tier j The output of a neuron; is the activation function; n Indicates the number of neurons in the previous layer; Indicates l Tier i Input to j The weights of the neurons; express l- 1st floor j The output of a neuron; Indicates j The bias of each neuron; Subsequently, the output value of the linear transformation is transformed nonlinearly through a nonlinear activation function as follows: ; in, Indicates L- The activation value vector of layer 1; represents the activation function; express L- Output value of layer 1; Finally, the predicted value of the electric vehicle's driving range is output in the output layer, and the calculation formula is as follows: ; in, Represents the weight matrix of the last layer; It is L- The activation value vector of layer 1; is the bias term of the output layer; is the final output, i.e. the predicted driving range.

[0013] Compared with the prior art, the present invention has the following beneficial effects: 1. Compared with the traditional electric vehicle range prediction model, the present invention has been improved. It not only considers the impact of energy consumption on the electric vehicle range, but also considers the impact of available energy on the range, including factors such as temperature and battery degradation, thereby improving the prediction accuracy of the electric vehicle range.

[0014] 2. In order to solve the complexity problem existing in the single-layer model, the present invention adopts a hierarchical prediction model to transform the electric vehicle range prediction problem into the electric vehicle energy consumption rate and actual energy coefficient prediction problem, so that each module can focus on the completion of subtasks, improve the flexibility and scalability of the model, reduce development time and cost, and greatly improve the accuracy of electric vehicle range prediction.

[0015] 3. The present invention adopts a driving range prediction model driven by data and mechanism synergistically, which not only overcomes the shortcomings of physical models that are difficult to capture complex nonlinear relationships and lack of real-time prediction, but also enhances the interpretability of the model through the introduction of physical mechanisms, reduces dependence on large-scale data, effectively avoids the problem of model overfitting, enhances user trust, and is conducive to the promotion of electric vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flow chart of the present invention.

[0017] Figure 2 This is a comparison chart of the predicted driving range of the test vehicles.

[0018] Figure 3 This is a real-time prediction effect diagram for the test vehicle's driving range. DETAILED DESCRIPTION

[0019] 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 used to limit the present invention.

[0020] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0021] like Figure 1 As shown, a knowledge data collaboratively driven hierarchical prediction method for electric vehicle driving range is provided by an embodiment of the present invention. The method realizes the prediction of driving range based on a hierarchical prediction model. The hierarchical prediction model includes an input layer, an upper modular estimation layer and a lower information fusion layer. The upper modular estimation layer includes an energy consumption rate prediction module and an actual energy coefficient prediction module. The hierarchical prediction model based on real-world data is implemented by a software system, running on an Ubuntu 18.04 operating system, and uses the PyTorch 2.3.1 package to construct the upper modular estimation layer and the lower information fusion layer.

[0022] The method comprises: 1) The input layer collects input features, including historical driving data, vehicle speed, battery temperature, and battery status, providing a basis for the operation of the subsequent upper modular estimation layer; The key characteristic variables that affect the energy consumption rate include the temperature in the heat release phase, idle speed ratio, idle energy consumption rate, vehicle speed, start-stop SOC and other information. This information is used as the input features of the energy consumption rate prediction module.

[0023] The key characteristic variables that affect the actual energy coefficient include temperature, total current, total voltage, and mileage data during the charging phase. This information is used as input features of the actual energy coefficient prediction module.

[0024] 2) Mechanism analysis: According to the mechanism definition, the driving range is the ratio of available energy to energy consumption rate, and the available energy is related to the actual energy coefficient. The formula is expressed as follows: ; ; in, RDR is the driving range of an electric vehicle, RE battery is the available energy of the electric vehicle, E cos is the energy consumption rate of the electric vehicle, SOC start is the initial battery state of charge value of the electric vehicle, SOC end is the terminated battery state-of-charge value, C is the actual energy coefficient of the electric vehicle. It can be seen that the driving range of an electric vehicle is closely related to its energy consumption rate and actual energy coefficient.

[0025] Therefore, on this basis, the upper modular estimation layer uses the mechanism knowledge (physical knowledge) during mechanism analysis to transform the problem of predicting the driving range of electric vehicles into the problem of predicting the energy consumption rate and actual energy coefficient of electric vehicles, thereby achieving accurate prediction of the energy consumption rate of electric vehicles.

[0026] a. Energy consumption rate refers to the amount of electricity consumed per kilometer during the driving process of an electric vehicle. The energy consumption rate prediction module uses a multi-layer perceptron (MLP) model to predict the energy consumption rate at the beginning of the journey, and a long short-term memory (LSTM) model to predict the energy consumption rate during the driving phase. The specific methods are as follows: First, the data of each discharge process is used to obtain the real energy consumption rate label; then, temperature (maximum temperature in the heat release stage, minimum temperature in the heat release stage), driving information (idle ratio, idle energy consumption rate, vehicle speed) and start and end SOC (starting SOC, ending SOC) are used as input features, and the mechanism knowledge (physical knowledge) used in the above mechanism analysis is used to calculate the energy consumption of each trip segment; then, the structure of switching between the MLP model and the LSTM model is used for prediction; at the beginning of the trip, due to insufficient data, the MLP model is used to call historical data for prediction. When sufficient data is collected (driving time ≥ 25 minutes), the MLP model autonomously switches to the LSTM model to learn the time series changes of the collected driving data, thereby improving the accuracy of the energy consumption rate prediction model; finally, hyperparameter tuning and K-fold cross validation are used to improve the prediction performance of the model.

[0027] b. The actual energy coefficient reflects the effectiveness of energy utilization in actual use of electric vehicles. The actual energy coefficient prediction module first uses each charging process data to obtain the real actual energy coefficient label. In order to enhance the prediction accuracy of the model, the Monte Carlo simulation method is used to improve the credibility of the training label to ensure that the model can more accurately reflect the actual energy utilization; then use temperature (highest temperature in the charging stage, lowest temperature in the charging stage), constant current charging segment data (total current, total voltage) and mileage data as input features to identify fast and slow charging, and predict the actual energy coefficient of fast charging and slow charging respectively; finally, use the LSTM model to learn the time series changes of the actual energy coefficient during vehicle driving. The specific method is as follows: ; in, The input gate determines how much new information is introduced into the memory cell; The forget gate controls how much old information is discarded; The output gate determines how much information is output from the memory cell to the next time step or the output layer of the network; is the time step t The candidate memory cell states; is the time step t The cell state is updated at each time step and consists of two parts: one is the result of multiplying the previous memory state by the forget gate, and the other is the candidate memory cell state AND input gate The result of multiplication; is the time step t The hidden state is the output of the LSTM unit at each time step, which is passed to the next layer or the next time step; is the activation function; is the input weight matrix corresponding to the input gate; is the input weight matrix corresponding to the forget gate; Is the input weight matrix corresponding to the output gate; is the input weight matrix corresponding to the candidate memory unit state; is the current time step t The input vector of is the recursive weight matrix of the input gate; is the recursive weight matrix of the forget gate; is the recursive weight matrix of the output gate; is the recursive weight matrix of the candidate memory unit; is the time step t- The hidden state of 1; is the bias vector of the input gate; is the bias vector of the forget gate; is the bias vector of the output gate; is the bias vector of the candidate memory cell state; is the time step t The forget gate activation value of is the time step t- 1's unit status; is the time step t The input gate activation value of is the time step t The output gate activation value of .

[0028] 3) According to the mechanism analysis, the driving range of electric vehicles is strongly correlated with the energy consumption rate and the actual energy coefficient. Therefore, the lower information fusion layer uses the energy consumption rate and the actual energy coefficient as the input features of this layer, and outputs the driving range of electric vehicles after passing through the MLP model. The MLP model includes an input layer, a hidden layer, and an output layer. The specific method is as follows: The input layer receives input features, which come from the outputs of the upper modular estimation layer, namely the energy consumption rate prediction module and the actual energy coefficient prediction module; assuming that the input vector from the upper modular estimation layer is X=[x1,x2,…,x n ], X is the entire input feature vector, including all input features; x1,x2,…,x n For each element in the input feature vector, that is, each feature or variable; n is the number of features; the input layer passes the feature vector to the first hidden layer; The neurons in the hidden layer receive the input features of the first input layer and perform linear transformation through weighted summation: ; in, Indicates l Tier j The output of a neuron; is the activation function; n Indicates the previous layer ( l- The number of neurons in layer 1; Indicates l Tier i Input to j The weights of the neurons; express l- 1st floor j The output of a neuron; Indicates j The bias of each neuron; Subsequently, the output value of the linear transformation is transformed nonlinearly through a nonlinear activation function as follows: ; in, Indicates L- The activation value vector of layer 1; represents the activation function; express L- Output value of layer 1; Finally, the predicted value of the electric vehicle's driving range is output in the output layer, and the calculation formula is as follows: ; in, Represents the weight matrix of the last layer; It is L- The activation value vector of layer 1; is the bias term of the output layer; is the final output, i.e. the predicted driving range.

[0029] Embodiment 1, training and verification; The collected real-world driving data of electric vehicles are divided into training, validation, and test sets in chronological order. Specifically, the first 90% of the data is used for the training and validation sets to train the model parameters and adjust the hyperparameters, and the last 10% of the data is used for the test set to evaluate the model performance. The model is built and trained based on the PyTorch framework, and all experiments are conducted on a PC server equipped with an Intel(R) Core(TM) i7-13700KF 3.40 GHz processor and a 12 GB NVIDIA GeForce RTX 4070Ti GPU.

[0030] To prevent overfitting of the deep learning model, the Dropout method with a probability of 0.3 was used and applied between the convolutional layer, the recurrent layer and the fully connected layer. Adam was selected as the optimizer, and the mean square error (MSE) was used as the loss function. After adjusting the hyperparameters on the validation set, the time step was finally determined to be 25, the batch size was 64, and the bidirectional LSTM was configured with 64 hidden neurons to fully learn the temporal features. The entire model training was carried out for 30 cycles, and the model evaluation indicators used were the mean absolute percentage error (MAPE) and the coefficient of determination ( R 2 ) to conduct a comprehensive and quantitative analysis of the experimental results. MAPE (mean absolute percentage error) is a statistical indicator used to measure the accuracy of the model, mainly used to evaluate the deviation between the model's predicted value and the actual value. MAPE represents the average percentage of prediction error and can intuitively reflect the accuracy of the prediction result relative to the actual result. The smaller its value, the higher the prediction accuracy of the model. R 2 A measure of how well the variability predicted by the model matches the actual variability in the data. R 2The value of is usually between 0 and 1. The closer the value is to 1, the better the model explains the variation of the data. R 2 The specific calculation formula is as follows: ; ; in, is the index variable used to iterate over each observation; n is the total number of observations; is the actual driving range of electric vehicles, is the driving range predicted by the hierarchical prediction model.

[0031] Use the trained model to infer the test set, and the comparison chart of the test vehicle mileage prediction is as follows Figure 2 As shown in the figure, it can be seen that the hierarchical prediction model R 2 The closest to 1, the smallest MAPE. The real-time prediction effect of the test vehicle's driving range is shown in the figure below: Figure 3 As shown in the figure, it can be seen that the hierarchical prediction model shows excellent prediction results both before and after switching. The above results show the effectiveness and versatility of the hierarchical prediction model. Cars 1-4 are predicted using the hierarchical prediction model, the End-to-End model, and the CLTC cycle, respectively. The relevant evaluation indicators of the model test set are shown in Tables 1, 2, and 3.

[0032] Table 1. Prediction indicators for different test vehicles R 2

[0033] Table 2. MAPE prediction indicators for different test vehicles

[0034] Table 3. Different indicators for real-time prediction of driving range of test vehicles

[0035] As can be seen from Table 1, the prediction indicators of the hierarchical prediction model for cars 1-4 R 2 It is closer to 1, which shows that the hierarchical prediction model explains the variation of the data better. As can be seen from Table 2, the MAPE value of the hierarchical prediction model prediction index of cars 1-4 is smaller, which shows that the prediction accuracy of the hierarchical prediction model is higher. As can be seen from Table 3, the real-time prediction result index of the hierarchical prediction model is R 2The closer it is to 1, the smaller the value of MAPE is, which indicates that the hierarchical forecasting model can better explain the variability of the data in real-time forecasting and has higher prediction accuracy.

[0036] In summary, the hierarchical prediction model is mainly based on data-driven and the mechanism knowledge (physical knowledge) used in the above-mentioned mechanism analysis, and adopts a hierarchical architecture, which can be quickly trained and predicted in real time with limited computing resources. In addition, the model uses K-fold cross-validation to improve the robustness of the model, and improves the prediction effect through hyperparameter tuning, so that it has higher accuracy and better generalization than other single-layer or single models, so it can accurately estimate the actual driving range of electric vehicles within the full range of working conditions. It can be seen that the knowledge data collaboratively driven hierarchical prediction method for electric vehicle driving range designed by the present invention for the problem of electric vehicle driving range prediction can accurately predict the driving range of electric vehicles in real time.

[0037] The above are only preferred embodiments of the present invention. It should be pointed out that, for those skilled in the art, several modifications and improvements can be made without departing from the concept of the present invention. These should also be regarded as the protection scope of the present invention. These will not affect the effect of the implementation of the present invention and the practicality of the patent.

Claims

1. A knowledge data collaboratively driven electric vehicle driving range hierarchical prediction method, characterized in that: The following steps are involved: Step S1, collecting input features, including historical driving data, vehicle speed, battery temperature and battery status; Step S2, converting the problem of predicting the driving range of the electric vehicle into the problem of predicting the energy consumption rate and actual energy coefficient of the electric vehicle; Step S3: combining the prediction results of the energy consumption rate and the actual energy coefficient to generate a predicted value of the driving range of the electric vehicle.

2. The knowledge data collaboratively driven electric vehicle driving range hierarchical prediction method according to claim 1 is characterized in that: The specific process of step S2 is as follows: Step 21: Use the MLP model to predict the energy consumption rate at the beginning of the trip, and use the LSTM model to predict the energy consumption rate during the driving phase. Step 22: First, use each charging process data to obtain the real actual energy coefficient label, and use the Monte Carlo simulation method to improve the credibility of the label; then use the temperature, constant current charging segment data and mileage data as input features to perform fast and slow charging identification, and perform fast charging actual energy coefficient prediction and slow charging actual energy coefficient prediction respectively; Finally, the LSTM model is used to learn the time series changes of the actual energy coefficient during vehicle driving.

3. The knowledge data collaboratively driven electric vehicle driving range hierarchical prediction method according to claim 2 is characterized in that: The specific method of step 21 is: first, use each discharge process data to obtain the real energy consumption rate label; then use temperature, driving information and start and end SOC as input features to calculate the energy consumption of each trip segment; then use the structure of switching between MLP model and LSTM model to make predictions; At the beginning of the journey, due to insufficient data, the MLP model is used to call historical data for prediction. When the amount of data reaches the preset threshold, the MLP model autonomously switches to the LSTM model to learn the time series changes of the collected driving data. Finally, hyperparameter tuning and K-fold cross-validation are used to improve the prediction performance of the model.

4. The knowledge data collaboratively driven electric vehicle driving range hierarchical prediction method according to claim 2 is characterized in that: The specific method of step 22 is as follows: ； in, is the input gate; For the Gate of Oblivion; is the output gate; is the time step t The candidate memory cell states; is the time step t The unit status; is the time step t The hidden state of is the activation function; is the input weight matrix corresponding to the input gate; is the input weight matrix corresponding to the forget gate; Is the input weight matrix corresponding to the output gate; is the input weight matrix corresponding to the candidate memory unit state; is the current time step t The input vector of is the recursive weight matrix of the input gate; is the recursive weight matrix of the forget gate; is the recursive weight matrix of the output gate; is the recursive weight matrix of the candidate memory unit; is the time step t- The hidden state of 1; is the bias vector of the input gate; is the bias vector of the forget gate; is the bias vector of the output gate; is the bias vector of the candidate memory cell state; is the time step t The forget gate activation value of is the time step t- 1's unit status; is the time step t The input gate activation value of is the time step t The output gate activation value of .

5. The knowledge data collaboratively driven electric vehicle driving range hierarchical prediction method according to claim 2 is characterized in that: The specific method of step S3 is as follows: Combining the prediction results of the energy consumption rate and the actual energy coefficient, the predicted value of the driving range of the electric vehicle is output through the MLP model, and the MLP model includes an input layer, a hidden layer and an output layer; The input layer receives input features. Assume that the input vector is , X is the entire input feature vector, including all input features; x1,x2,…,x n For each element in the input feature vector, that is, each feature or variable; n is the number of features; the input layer passes the feature vector to the first hidden layer; The neurons in the hidden layer receive the input features of the first input layer and perform linear transformation through weighted summation: ； in, Indicates l Tier j The output of a neuron; is the activation function; n Indicates the number of neurons in the previous layer; Indicates l Tier i Input to j The weights of the neurons; express l- 1st floor j The output of a neuron; Indicates j The bias of each neuron; Subsequently, the output value of the linear transformation is transformed nonlinearly through a nonlinear activation function as follows: ； in, Indicates L- The activation value vector of layer 1; represents the activation function; express L- Output value of layer 1; Finally, the predicted value of the electric vehicle's driving range is output in the output layer, and the calculation formula is as follows: ； in, Represents the weight matrix of the last layer; It is L- The activation value vector of layer 1; is the bias term of the output layer; is the final output, i.e. the predicted driving range.

Images