An electric vehicle driving range estimation method, device, equipment and medium

Abstract

The application relates to the technical field of electric vehicles, and provides an electric vehicle driving range estimation method, device, equipment and medium, which comprises the following steps: obtaining historical travel data and current travel data of an electric vehicle, and a first weight and a second weight; obtaining a plurality of first working condition types according to the historical travel data and the first weight, and obtaining a plurality of second working condition types according to the current travel data and the second weight; if the occurrence probability of a working condition type in the plurality of first working condition types is greater than a preset high occurrence probability threshold, the corresponding working condition type is determined as a high occurrence probability working condition type, and a second working condition type corresponding to the high occurrence probability working condition type in the plurality of second working condition types is determined as a future working condition type; and the current travel data and the future working condition type are input into a pre-trained driving range estimation model to obtain a current travel electric vehicle driving range estimation result. Through analysis of the future working condition, the problem that the future working condition is greatly different from the current working condition can be avoided.

Description

Technical Field

[0001] This application relates to the field of electric vehicle technology, specifically to a method, apparatus, device, and medium for estimating the driving range of an electric vehicle. Background Technology

[0002] Lithium-ion batteries used in electric vehicles generate electricity through an electrochemical reaction in which lithium atoms oxidize and move to the other electrode in the electrolyte to combine with electrons, thus driving the vehicle. Based on this principle, cell energy density, cell quantity, battery management system, electric drive system, and overall vehicle structure are fundamental conditions for ensuring the driving range of electric vehicles.

[0003] In actual driving, factors such as user driving behavior, external temperature and precipitation, road friction, and wind can all affect battery performance, resulting in a difference between actual energy consumption and theoretical energy consumption, which in turn affects the driving range of electric vehicles.

[0004] Chinese patent CN109094376A discloses a method and apparatus for calculating driving range. This method uses a gradient descent approach to fit the error between the actual and displayed mileage values ​​based on the vehicle's driving range performance over multiple cycles, gradually reducing the error. While this approach can estimate driving range based on a large amount of historical average vehicle data, it does not consider the impact of real-world user behavior, environmental changes, and fluctuations in battery properties on driving range.

[0005] Chinese patent CN105301508A discloses a radial basis function (RBF) neural network method for estimating the driving range of electric vehicles. This method involves inputting voltage, current, and speed data into the RBF neural network based on power battery data and ABS control unit data. Clustering is used to determine the hidden layer centers, and finally, least squares method is applied to calculate the driving range. While this approach addresses the issue of previous patents only considering historical average performance, the parameters of this type of neural network are obtained empirically, resulting in poor model interpretability and neglecting the influence of external factors.

[0006] Chinese patent CN110667434A discloses an adaptive driving condition method and system for estimating the driving range of a pure electric vehicle. This method determines the real-time driving condition based on the current road environment and traffic conditions, and uses a backpropagation neural network to predict the energy consumption data under this real-time condition based on historical driving energy consumption data, using this prediction as the input variable for the remaining driving range. While this solution considers the influence of driving conditions, it uses real-time driving conditions to estimate energy consumption but does not fully account for the potential significant differences between future and current driving conditions.

[0007] Therefore, how to conduct a comprehensive analysis of future operating conditions and how to improve the interpretability of the driving range estimation model are urgent problems that need to be solved. Summary of the Invention

[0008] In view of the shortcomings of the prior art described above, the purpose of this application is to provide a method, apparatus, device and medium for estimating the driving range of electric vehicles, so as to solve the problems in the prior art of how to comprehensively analyze future operating conditions and how to improve the interpretability of the driving range estimation model.

[0009] To achieve the above and other related objectives, this application provides a method for estimating the driving range of an electric vehicle, the method comprising:

[0010] Acquire historical and current trip data of electric vehicles, as well as a first weight for trip energy consumption corresponding to the historical trip data and a second weight for trip energy consumption corresponding to the current trip data;

[0011] Multiple first working condition types are obtained based on the historical travel and the first weight, and multiple second working condition types are obtained based on the current travel and the second weight;

[0012] If the probability of occurrence of a certain working condition type among the plurality of first working condition types is greater than a preset high probability threshold, then the corresponding working condition type is determined to be a high probability working condition type, and the second working condition type corresponding to the high probability working condition type among the plurality of second working condition types is determined to be a future working condition type.

[0013] The current trip data and future operating condition type are input into a pre-trained range estimation model to obtain the estimated range of the electric vehicle for the current trip.

[0014] In one embodiment of this application, the plurality of first driving condition types include high-speed smooth driving, high-speed aggressive driving, urban smooth driving, and urban aggressive driving;

[0015] The multiple second driving conditions include high-speed smooth driving, high-speed aggressive driving, urban smooth driving, and urban aggressive driving.

[0016] In one embodiment of this application, the historical travel data includes historical battery data, historical motor operating data, and historical external environment data; the current travel data includes current battery data, current motor operating data, and current external environment data. Obtaining a first weight for travel energy consumption corresponding to the historical travel data and a second weight for travel energy consumption corresponding to the current travel data includes:

[0017] The historical trip data and current trip data are normalized to obtain normalized historical trip data and current trip data;

[0018] Obtain the first Spearman correlation coefficients of historical battery data, historical motor operating data, and historical external environment data corresponding to the energy consumption of the trip from the normalized historical trip data, and assign a first weight to the historical trip data based on the first Spearman correlation coefficients;

[0019] Obtain the second Spearman correlation coefficient corresponding to the energy consumption of the current trip in the normalized current trip data, including the current battery data, current motor operating data, and current external environment data, and assign a second weight to the current trip data based on the second Spearman correlation coefficient.

[0020] In one embodiment of this application, multiple first working condition types are obtained based on the historical trip and the first weight, and multiple second working condition types are obtained based on the current trip and the second weight, including:

[0021] The normalized historical travel data is multiplied by the first weight to obtain multiple first initial working condition types, and the multiple first initial working condition types are aggregated and classified according to a preset distance clustering method to obtain multiple first working condition types.

[0022] The normalized current travel data is multiplied by the second weight to obtain multiple second initial working condition types. The multiple second initial working condition types are then aggregated and classified according to a preset distance clustering method to obtain multiple second working condition types.

[0023] In one embodiment of this application, the historical trip data further includes historical driving range data. Before inputting the current trip data and future operating condition type into the pre-trained driving range estimation model, the method further includes:

[0024] The historical trip data and the first operating condition type are input into the pre-built initial driving range estimation model;

[0025] The initial driving range estimation model is trained using the historical travel data and the first operating condition type to update the model weights and thresholds of the initial driving range estimation model, thereby obtaining the pre-trained driving range estimation model.

[0026] In one embodiment of this application, the pre-built initial driving range estimation model includes a radial basis function neural network model, and the model weights include a first model weight, a second model weight, and a third model weight. The radial basis function neural network model includes:

[0027] The input layer is a signal source node used to receive the historical data and the first operating condition type;

[0028] The hidden layer has a hidden unit activation function that is a radial basis function. The radial basis function is used to spatially map the historical travel data and the first working condition type through the first model weight and the second model weight to obtain the spatially mapped historical travel data and the first working condition type.

[0029] The output layer is used to linearly weight the spatially mapped historical travel data and the first working condition type according to the third weight and then output the result.

[0030] In one embodiment of this application, training the initial driving range estimation model using the historical trip data and the first operating condition type includes:

[0031] Obtain the threshold and network error of the radial basis function neural network model;

[0032] The historical travel data and the first working condition type are binary encoded to obtain the population;

[0033] Set the model weights and thresholds as decision variables, and set the network error as the fitness value;

[0034] Based on the decision variables and fitness values, the population is subjected to multiple analogical natural selections to retain optimal solutions and crossover mutations to generate new solutions. If the new solutions meet the preset training termination conditions, the optimal model weights and optimal thresholds are obtained. The optimal model weights and optimal thresholds are used to update the initial driving range estimation model.

[0035] In one embodiment of this application, an electric vehicle range estimation device is also provided, the device comprising:

[0036] The information acquisition module is used to acquire historical travel data and current travel data of electric vehicles, as well as a first weight for travel energy consumption corresponding to the historical travel data and a second weight for travel energy consumption corresponding to the current travel data;

[0037] The working condition classification module is used to obtain multiple first working condition types based on the historical trip and the first weight, and to obtain multiple second working condition types based on the current trip and the second weight;

[0038] The future working condition type determination module is used to determine the corresponding working condition type as a high-probability working condition type if the probability of occurrence of a certain working condition type among the plurality of first working condition types is greater than a preset high-probability threshold, and to determine the second working condition type among the plurality of second working condition types that corresponds to the high-probability working condition type as the future working condition type.

[0039] The driving range estimation module is used to input the current trip data and future operating condition type into a pre-trained driving range estimation model to obtain the driving range estimation result of the electric vehicle for the current trip.

[0040] In one embodiment of this application, an electronic device is also provided, the electronic device comprising:

[0041] One or more processors;

[0042] A storage device for storing one or more programs that, when executed by one or more processors, cause the electronic device to implement the electric vehicle range estimation method as described above.

[0043] In one embodiment of this application, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a computer's processor, causes the computer to perform the electric vehicle range estimation method as described above.

[0044] The beneficial effects of this invention are:

[0045] First, historical and current travel data of the electric vehicle are acquired, along with a first weight corresponding to travel energy consumption for the historical travel data and a second weight corresponding to travel energy consumption for the current travel data. Multiple first operating condition types are obtained based on the historical travel data and the first weight, and multiple second operating condition types are obtained based on the current travel data and the second weight. If the probability of occurrence of a certain operating condition type among the multiple first operating condition types is greater than a preset high probability threshold, the corresponding operating condition type is determined as a high probability operating condition type, and the second operating condition type corresponding to the high probability operating condition type among the multiple second operating condition types is determined as a future operating condition type. The current travel data and future operating condition types are input into a pre-trained range estimation model to obtain the current travel range estimation result of the electric vehicle. In this invention, multiple operating condition types are obtained based on historical travel data and their corresponding weights, and current travel data and their corresponding weights. The future operating condition type is determined based on the probability of occurrence of the first operating condition type corresponding to the historical travel data. By comprehensively analyzing the future operating condition types and estimating the driving range using the future operating condition types, the problem of a large discrepancy between future and current operating conditions can be avoided, thereby improving the accuracy of the driving range estimation result.

[0046] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0047] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. In the drawings:

[0048] Figure 1 This is a schematic diagram illustrating the implementation environment of an electric vehicle range estimation method, as shown in an exemplary embodiment of this application.

[0049] Figure 2 This is a flowchart illustrating an exemplary embodiment of the method for estimating the driving range of an electric vehicle, as shown in this application.

[0050] Figure 3 This is a schematic diagram of the comprehensive analysis logic of the working conditions shown in an exemplary embodiment of this application;

[0051] Figure 4 This is a schematic diagram of a radial basis function neural network structure shown in an exemplary embodiment of this application;

[0052] Figure 5 This is a schematic diagram of the genetic optimization network model illustrated in an exemplary embodiment of this application;

[0053] Figure 6 This is a block diagram illustrating an electric vehicle range estimation device according to an exemplary embodiment of this application;

[0054] Figure 7 A schematic diagram of the structure of a computer system suitable for an electronic device according to an embodiment of this application is shown. Detailed Implementation

[0055] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0056] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0057] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0058] First, it's important to note that with the increasing popularity of electric vehicles, driving range has become a key focus for car owners when purchasing and using them. The driving range of an electric vehicle is the total distance it can continuously travel under current battery charge, driving conditions, and environmental conditions. Most existing methods for estimating driving range directly estimate average energy consumption based on historical speed and mileage, without considering the impact of changes in driving behavior and fluctuations in battery properties. Some technologies use deep learning models for range estimation, but deep learning models obtained directly from preset parameters have poor interpretability and do not consider the influence of external environmental factors on driving range. Furthermore, while some patents consider estimating driving range based on operating conditions, this is only based on real-time or historical average operating conditions, lacking comprehensive analysis of future operating conditions, resulting in low reliability and accuracy of the estimated results.

[0059] The following explains the technical names used in this application:

[0060] A radial basis function (RBF) neural network is a neural network that uses RBFs as activation functions. A RBF is a real-valued function whose value depends only on its distance from the origin, or on its distance from any point. The distance is typically Euclidean distance, but other distance functions can also be used. Any function that satisfies this property is called a radial basis function.

[0061] Genetic Algorithm (GA) is a computational model that simulates the biological evolutionary process based on natural selection and genetic mechanisms, as described in Darwin's theory of evolution. It's a method for searching for optimal solutions by simulating natural evolution. This algorithm uses mathematical methods and computer simulations to transform the problem-solving process into processes similar to crossover and mutation of chromosomes and genes in biological evolution. When solving complex combinatorial optimization problems, it typically achieves better optimization results faster than some conventional optimization algorithms.

[0062] Figure 1 This is a schematic diagram illustrating the implementation environment of an electric vehicle range estimation method, as shown in an exemplary embodiment of this application. Figure 1As shown, the implementation environment includes a terminal device 101, a cloud server 102, and a data acquisition device 103. The terminal device 101 acquires vehicle travel data collected by the data acquisition device 103 and estimates the remaining driving range. The estimated result is then transmitted to the cloud server 102 for storage or further application.

[0063] in, Figure 1 The terminal device 101 shown may be, for example, an in-vehicle terminal, a tablet computer, a laptop computer, or any other terminal device capable of carrying a pre-trained electric vehicle range estimation model, but is not limited thereto. Figure 1 The cloud server 102 shown can be, for example, a server cluster or distributed system built on multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. There are no restrictions on this. Figure 1 The data acquisition device 103 shown may be, for example, a sensor, a BMS (Battery Management System), or a motor status monitoring module; this application does not impose any specific limitations on it.

[0064] like Figure 1 As shown, the electric vehicle range estimation method in this embodiment can be implemented, for example, through an on-board terminal configured with a pre-trained electric vehicle range estimation model, by performing the following steps: acquiring historical travel data and current travel data of the electric vehicle, as well as a first weight corresponding to travel energy consumption for the historical travel data and a second weight corresponding to travel energy consumption for the current travel data; obtaining multiple first operating condition types based on the historical travel data and the first weight, and obtaining multiple second operating condition types based on the current travel data and the second weight; if the probability of occurrence of a certain operating condition type among the multiple first operating condition types is greater than a preset high probability threshold, then determining the corresponding operating condition type as a high probability operating condition type, and determining the second operating condition type among the multiple second operating condition types corresponding to the high probability operating condition type as a future operating condition type; inputting the current travel data and the future operating condition type into the pre-trained range estimation model to obtain the current travel electric vehicle range estimation result.

[0065] To address the problems of existing driving range estimation methods that do not consider environmental changes, battery property fluctuations, future operating conditions, and the poor interpretability of deep learning models, embodiments of this application propose an electric vehicle driving range estimation method, an electric vehicle driving range estimation device, an electronic device, a computer-readable storage medium, and a computer program product, which will be described in detail below.

[0066] Please see Figure 2 , Figure 2 This is a flowchart illustrating an exemplary embodiment of an electric vehicle range estimation method, which can be applied to... Figure 1 The implementation environment is shown. It should be understood that this method can also be applied to other exemplary implementation environments and specifically executed by devices in other implementation environments. This embodiment does not limit the implementation environment to which the method is applicable.

[0067] like Figure 2 As shown, in an exemplary embodiment, the electric vehicle range estimation method includes at least steps S210 to S240, which are described in detail below:

[0068] In step S210, historical travel data and current travel data of the electric vehicle are obtained, as well as a first weight for travel energy consumption corresponding to historical travel data and a second weight for travel energy consumption corresponding to current travel data.

[0069] First, it should be noted that the historical trip data in this embodiment includes historical battery data, historical motor operating data, and historical external environment data. Taking the previous trip as an example, in the previous trip, the vehicle's battery data was obtained through the battery management system, the motor operating data was obtained through the battery status management module, and the external environment data was obtained through sensors. The current trip data includes current battery data, current motor operating data, and current external environment data. Battery data may include, for example, remaining battery power, battery fluctuation information, and battery health status; motor operating data may include, for example, motor speed and acceleration, and can also be understood as driving data; external environment data may include, for example, external temperature and humidity. It should be understood that this embodiment does not limit the types of data in the historical trip data and the current trip data, and appropriate additions can be made according to actual needs.

[0070] It should also be noted that the first weight and the second weight in the embodiments of this application are the relationship parameters between the travel data and the travel energy consumption obtained in advance through analysis and calculation, that is, the degree of influence of the travel data on the travel energy consumption. Different types of travel data can be regarded as the influence factors of different degrees of influence on travel energy consumption.

[0071] In step S220, multiple first working condition types are obtained based on historical travel and first weight, and multiple second working condition types are obtained based on current travel and second weight.

[0072] In one embodiment of this application, multiple first driving condition types include high-speed smooth driving, high-speed aggressive driving, urban smooth driving, and urban aggressive driving; multiple second driving condition types include high-speed smooth driving, high-speed aggressive driving, urban smooth driving, and urban aggressive driving. It should be noted that the embodiments of this application only illustrate the first and second driving condition types as examples, and do not imply that the types are limited to these, but can be set according to actual circumstances.

[0073] For example, according to the travel window in the electric vehicle range estimation system in terminal device 101, the K-distance clustering method is used to divide the weights and travel data into multiple operating condition types such as high-speed smooth driving, high-speed aggressive driving, urban smooth driving, and urban aggressive driving.

[0074] In step S230, if the probability of occurrence of a certain working condition type among the multiple first working condition types is greater than a preset high probability threshold, then the corresponding working condition type is determined to be a high probability working condition type, and the second working condition type corresponding to the high probability working condition type among the multiple second working condition types is determined to be a future working condition type.

[0075] For example, obtain the probability of each working condition type occurring among multiple first working condition types, and determine the probability P of a certain working condition type occurring among the multiple first working condition types. Ci Is it greater than the preset high probability threshold? If the probability P of a certain working condition type occurs... Ci If the probability of occurrence exceeds a preset high-probability threshold, the operating condition type can be considered a high-probability operating condition type. Then, based on this high-probability operating condition type, the corresponding operating condition type from multiple second operating condition types is set as the future operating condition type. In this embodiment, given historical travel data, a higher probability of occurrence of a certain operating condition indicates a higher probability of occurrence of that operating condition in the future scenario corresponding to the current travel data. In the comprehensive analysis of future operating condition types, battery data, motor operating data, and external environment data are used as influencing factors. Compared to the existing method of directly estimating the driving range based on the vehicle's historical average data, the estimated result is closer to the actual driving range under the given driving environment, with smaller errors.

[0076] In one embodiment of this application, after obtaining the first working condition type and the second working condition type, a cross-trip working condition table of historical trips and current trips can be created, and the probability of occurrence of each working condition type in multiple first working condition types can be added to the cross-trip working condition table to obtain the corresponding future working condition type in multiple second working condition types.

[0077] In step S240, the current trip data and future operating condition type are input into the pre-trained driving range estimation model to obtain the driving range estimation result of the electric vehicle for the current trip.

[0078] For example, the pre-trained driving range estimation model is a pre-trained radial basis function (RBF) neural network model, and a genetic algorithm is used to address the low interpretability problem of RBF neural network models. Current trip data and future driving condition type C are used as input features X, which are then input into the pre-trained RBF neural network model RBF_Model. The genetic algorithm iterates multiple times, using weights and thresholds as decision variables and network error as the fitness value, until the model meets the termination condition. Finally, a highly accurate genetically optimized network prediction model RBF_GA_Model and the driving range estimation result Y are obtained.

[0079] As can be seen from steps S210 to S240 above, the solution proposed in this embodiment obtains multiple operating condition types based on historical travel data and their corresponding weights, current travel data and their corresponding weights, and determines the future operating condition type based on the occurrence probability of the first operating condition type corresponding to the historical travel data. By comprehensively analyzing the future operating condition types and estimating the driving range based on the future operating condition types, the problem of a large difference between the future operating conditions and the current operating conditions can be avoided, thereby improving the accuracy of the driving range estimation results.

[0080] In one embodiment of this application, Figure 2 Step S210, as shown, involves obtaining the first weight corresponding to the historical trip data and the second weight corresponding to the current trip data, which includes the following steps:

[0081] The historical trip data and current trip data are normalized to obtain normalized historical trip data and current trip data;

[0082] Obtain the first Spearman correlation coefficients of historical battery data, historical motor operating data, and historical external environment data corresponding to the energy consumption of the trip from the normalized historical trip data, and assign a first weight to the historical trip data based on the first Spearman correlation coefficients;

[0083] Obtain the second Spearman correlation coefficient corresponding to the energy consumption of the current trip in the normalized current trip data, including the current battery data, current motor operating data, and current external environment data, and assign a second weight to the current trip data based on the second Spearman correlation coefficient.

[0084] In one embodiment of this application, multiple first working condition types are obtained based on historical travel and a first weight, and multiple second working condition types are obtained based on the current travel and a second weight, including:

[0085] The normalized historical travel data is multiplied by the first weight to obtain multiple first initial working condition types, and the multiple first initial working condition types are aggregated and classified according to a preset distance clustering method to obtain multiple first working condition types.

[0086] The normalized current travel data is multiplied by the second weight to obtain multiple second initial working condition types. The multiple second initial working condition types are then aggregated and classified according to a preset distance clustering method to obtain multiple second working condition types.

[0087] In one embodiment of the application, the operating condition comprehensive analysis logic is described in detail. See also Figure 3 , Figure 3 This is a schematic diagram illustrating the logic of comprehensive operating condition analysis in an exemplary embodiment of this application. To define future operating condition types, a comprehensive analysis of real operating conditions is performed based on historical data; to determine the future operating condition type corresponding to the current trip data, a cross-trip operating condition probability table is drawn. Acquired data includes BMS battery data, motor driving data, and sensor external environment data. The above data is normalized, and weights are assigned according to the Spearman correlation coefficient between each data type and trip energy consumption. After multiplying the normalized data by the weights, all operating conditions are aggregated into k categories using the k-distance clustering method. For example, 5 to 10 categories can be divided according to business needs, and the value of k can be determined based on the profile coefficient. The operating condition type and occurrence probability P of the previous trip are then... Ci Starting from this point, a cross-trip operating condition probability table is drawn to distinguish different regions and seasons to calculate the future operating condition category C corresponding to the current trip data. The future operating condition category C is then used as the input to the electric vehicle range estimation model.

[0088] In one embodiment of this application, the historical trip data further includes historical driving range data. Before inputting the current trip data and future operating condition type into the pre-trained driving range estimation model, the following steps are also included:

[0089] The historical trip data and the first operating condition type are input into the pre-built initial driving range estimation model;

[0090] The initial driving range estimation model is trained using the historical travel data and the first operating condition type to update the model weights and thresholds of the initial driving range estimation model, thereby obtaining the pre-trained driving range estimation model.

[0091] In one embodiment of this application, the pre-constructed initial driving range estimation model includes a radial basis function neural network model, and the model weights include a first model weight, a second model weight, and a third model weight. The radial basis function neural network model includes:

[0092] The input layer is a signal source node used to receive the historical data and the first operating condition type;

[0093] The hidden layer has a hidden unit activation function that is a radial basis function. The radial basis function is used to spatially map the historical travel data and the first working condition type through the first model weight and the second model weight to obtain the spatially mapped historical travel data and the first working condition type.

[0094] The output layer is used to linearly weight the spatially mapped historical travel data and the first working condition type according to the third weight and then output the result.

[0095] It should be noted that the radial basis function neural network model consists of three layers: the first layer is the input layer, which is the signal source node; the second layer is the hidden layer, the number of units of which is set as needed, and the activation function of the hidden units is the radial basis function, which is a non-negative nonlinear function that decays radially symmetric about the center point; the third layer is the output layer, which responds to the input pattern.

[0096] See Figure 4 , Figure 4 This is a schematic diagram of a radial basis function neural network (RBN) structure, illustrating an exemplary embodiment of this application. A RBN typically comprises three layers: an input layer, hidden layers, and an output layer. The input layer consists of signal source nodes and does not perform any transformation on the input information. The hidden layers perform a spatial mapping transformation on the input information, with the action function being a Gaussian function and the weights being C. ij d ij The output layer linearly weights the information output by the hidden layer neurons before outputting the final product. The action function is a linear function with weights w. ij .

[0097] In one embodiment of this application, the initial driving range estimation model is trained using historical trip data and a first operating condition type, including the following steps:

[0098] Obtain the threshold and network error of the radial basis function neural network model;

[0099] The historical travel data and the first working condition type are binary encoded to obtain the population;

[0100] Set the model weights and thresholds as decision variables, and set the network error as the fitness value;

[0101] Based on the decision variables and fitness values, the population is subjected to multiple analogical natural selections to retain optimal solutions and crossover mutations to generate new solutions. If the new solutions meet the preset training termination conditions, the optimal model weights and optimal thresholds are obtained. The optimal model weights and optimal thresholds are used to update the initial driving range estimation model.

[0102] In this embodiment, to address the issue of low interpretability in radial basis function neural network (RBN) models, a genetic algorithm, analogous to natural selection for preserving optimal solutions and crossover / mutation for generating new solutions, is used to optimize the model. The initial weights and thresholds of the network neurons are determined, and the run samples are encoded using a genetic algorithm. The network error is used as the fitness value for selection, crossover, and mutation until an optimal RBN model that meets the termination conditions is obtained.

[0103] For example, see Figure 5 , Figure 5 This is a schematic diagram illustrating the logic of a genetic optimization network model, as shown in an exemplary embodiment of this application. Based on the aforementioned neural network structure, genetic algorithm logic is added. Following the principle of "natural selection preserving optimal solutions and crossover mutation generating new solutions," a genetic optimization network prediction model RBF_GA_Model is constructed to increase the model's interpretability and improve its accuracy. The overall model logic is as follows: Figure 5 As shown: weights and thresholds are set as decision variables. Historical travel data and the first working condition type are binary encoded to obtain a population. The error of the basic radial basis function neural network model is used as the fitness value to evaluate the quality of travel samples and as the probability of selection in the genetic algorithm, helping the model to converge quickly. Based on the decision variables and fitness values, crossover coefficients and mutation coefficients are set, and the population is subjected to multiple analogical natural selection to retain the best solution and crossover mutation to generate new solutions. If the new solution meets the preset training termination condition, the optimal model weights and optimal thresholds are obtained. The optimal model weights and optimal thresholds are used to update the initial driving range estimation model. If the new solution does not meet the preset training termination condition, analogical natural selection is performed again to retain the best solution and crossover mutation to generate new solutions. The error of the model after updating the weights and thresholds is calculated. If the error meets the preset termination condition, the prediction result can be output according to the model. If the error does not meet the preset termination condition, the error is recalculated and it is determined whether the preset termination condition is met.

[0104] Figure 6 This is a block diagram illustrating an electric vehicle range estimation device according to an exemplary embodiment of this application. The device can be applied to... Figure 1 The implementation environment shown is not limited to this embodiment. This device can also be applied to other exemplary implementation environments and specifically configured in other devices. This embodiment does not limit the implementation environment to which the device is applicable.

[0105] like Figure 6As shown, this exemplary electric vehicle range estimation device includes:

[0106] The information acquisition module 601 is used to acquire historical travel data and current travel data of electric vehicles, as well as a first weight for travel energy consumption corresponding to the historical travel data and a second weight for travel energy consumption corresponding to the current travel data;

[0107] The working condition classification module 602 is used to obtain multiple first working condition types based on the historical trip and the first weight, and to obtain multiple second working condition types based on the current trip and the second weight;

[0108] The future working condition type determination module 603 is used to determine the corresponding working condition type as a high-probability working condition type if the probability of occurrence of a certain working condition type among the plurality of first working condition types is greater than a preset high-probability threshold, and to determine the second working condition type among the plurality of second working condition types that corresponds to the high-probability working condition type as the future working condition type.

[0109] The driving range estimation module 604 is used to input the current trip data and future operating condition type into a pre-trained driving range estimation model to obtain the driving range estimation result of the electric vehicle for the current trip.

[0110] In this exemplary electric vehicle range estimation device, multiple operating condition types are obtained based on historical travel data and their corresponding weights, current travel data and their corresponding weights, and the future operating condition type is determined based on the occurrence probability of the first operating condition type corresponding to the historical travel data. By comprehensively analyzing the future operating condition types and estimating the driving range using the future operating condition types, the problem of large discrepancies between the future operating conditions and the current operating conditions can be avoided, thereby improving the accuracy of the driving range estimation results.

[0111] It should be noted that the electric vehicle range estimation device and the electric vehicle range estimation method provided in the above embodiments belong to the same concept. The specific operation methods of each module and unit have been described in detail in the method embodiments and will not be repeated here. In practical applications, the road condition refresh device provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0112] Embodiments of this application also provide an electronic device, including: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the electric vehicle range estimation method provided in the above embodiments.

[0113] Figure 7 A schematic diagram of a computer system suitable for an electronic device according to an embodiment of this application is shown. It should be noted that... Figure 7 The computer system 700 of the electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0114] like Figure 7 As shown, the computer system 700 includes a Central Processing Unit (CPU) 701, which can perform various appropriate actions and processes based on programs stored in Read-Only Memory (ROM) 702 or programs loaded from storage portion 708 into Random Access Memory (RAM) 703, such as performing the methods described in the above embodiments. The RAM 703 also stores various programs and data required for system operation. The CPU 701, ROM 702, and RAM 703 are interconnected via a bus 704. An Input / Output (I / O) interface 705 is also connected to the bus 704.

[0115] The following components are connected to the I / O interface 705: an input section 706 including a keyboard, mouse, etc.; an output section 707 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the I / O interface 705 as needed. A removable medium 711, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 710 as needed so that computer programs read from it can be installed into the storage section 708 as needed.

[0116] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including a computer program for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 709, and / or installed from removable medium 711. When the computer program is executed by central processing unit (CPU) 701, it performs various functions defined in the system of this application.

[0117] It should be noted that the computer-readable medium shown in the embodiments of this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying a computer-readable computer program. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0118] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0119] The units described in the embodiments of this application can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself.

[0120] Another aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a computer's processor, causes the computer to perform the electric vehicle range estimation method as described above. This computer-readable storage medium may be included in the electronic device described in the above embodiments, or it may exist independently and not incorporated into the electronic device.

[0121] Another aspect of this application provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the electric vehicle range estimation method provided in the various embodiments described above.

[0122] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for estimating the driving range of an electric vehicle, characterized in that, The method includes: Acquire historical and current trip data of electric vehicles, as well as a first weight for trip energy consumption corresponding to the historical trip data and a second weight for trip energy consumption corresponding to the current trip data; Multiple first driving condition types are obtained based on the historical travel data and the first weight, and multiple second driving condition types are obtained based on the current travel data and the second weight. The multiple first driving condition types include highway smooth driving, highway aggressive driving, urban smooth driving, and urban aggressive driving. The multiple second driving condition types include highway smooth driving, highway aggressive driving, urban smooth driving, and urban aggressive driving. If the probability of occurrence of a certain working condition type among the plurality of first working condition types is greater than a preset high probability threshold, then the corresponding working condition type is determined to be a high probability working condition type, and the second working condition type corresponding to the high probability working condition type among the plurality of second working condition types is determined to be a future working condition type. The current trip data and future operating condition type are input into a pre-trained range estimation model to obtain the estimated range of the electric vehicle for the current trip.

2. The method for estimating the driving range of an electric vehicle according to claim 1, characterized in that, The historical travel data includes historical battery data, historical motor operating data, and historical external environment data. The current travel data includes current battery data, current motor operating data, and current external environment data. Obtaining the first weight of the historical travel data corresponding to travel energy consumption and the second weight of the current travel data corresponding to travel energy consumption includes: The historical trip data and current trip data are normalized to obtain normalized historical trip data and current trip data; Obtain the first Spearman correlation coefficients of historical battery data, historical motor operating data, and historical external environment data corresponding to the energy consumption of the trip from the normalized historical trip data, and assign a first weight to the historical trip data based on the first Spearman correlation coefficients; Obtain the second Spearman correlation coefficient corresponding to the energy consumption of the current trip in the normalized current trip data, including the current battery data, current motor operating data, and current external environment data, and assign a second weight to the current trip data based on the second Spearman correlation coefficient.

3. The method for estimating the driving range of an electric vehicle according to claim 2, characterized in that, Multiple first working condition types are obtained based on the historical travel data and the first weight, and multiple second working condition types are obtained based on the current travel data and the second weight, including: The normalized historical travel data is multiplied by the first weight to obtain multiple first initial working condition types, and the multiple first initial working condition types are aggregated and classified according to a preset distance clustering method to obtain multiple first working condition types. The normalized current travel data is multiplied by the second weight to obtain multiple second initial working condition types. The multiple second initial working condition types are then aggregated and classified according to a preset distance clustering method to obtain multiple second working condition types.

4. The method for estimating the driving range of an electric vehicle according to claim 2, characterized in that, The historical trip data also includes historical driving range data. Before inputting the current trip data and future operating condition type into the pre-trained driving range estimation model, the following steps are also included: The historical trip data and the first operating condition type are input into the pre-built initial driving range estimation model; The initial driving range estimation model is trained using the historical travel data and the first operating condition type to update the model weights and thresholds of the initial driving range estimation model, thereby obtaining the pre-trained driving range estimation model.

5. The method for estimating the driving range of an electric vehicle according to claim 4, characterized in that, The pre-built initial driving range estimation model includes a radial basis function neural network model, and the model weights include a first model weight, a second model weight, and a third model weight. The radial basis function neural network model includes: The input layer is a signal source node used to receive the historical travel data and the first working condition type; The hidden layer has a hidden unit activation function that is a radial basis function. The radial basis function is used to spatially map the historical travel data and the first working condition type through the first model weight and the second model weight to obtain the spatially mapped historical travel data and the first working condition type. The output layer is used to linearly weight the spatially mapped historical travel data and the first working condition type according to the third model weights and then output the result.

6. The method for estimating the driving range of an electric vehicle according to claim 5, characterized in that, The initial driving range estimation model is trained using the historical travel data and the first operating condition type, including: Obtain the threshold and network error of the radial basis function neural network model; The historical travel data and the first working condition type are binary encoded to obtain the population; Set the model weights and thresholds as decision variables, and set the network error as the fitness value; Based on the decision variables and fitness values, the population is subjected to multiple analogical natural selections to retain optimal solutions and crossover mutations to generate new solutions. If the new solutions meet the preset training termination conditions, the optimal model weights and optimal thresholds are obtained. The optimal model weights and optimal thresholds are used to update the initial driving range estimation model.

7. An electric vehicle range estimation device, characterized in that, The device includes: The information acquisition module is used to acquire historical travel data and current travel data of electric vehicles, as well as a first weight for travel energy consumption corresponding to the historical travel data and a second weight for travel energy consumption corresponding to the current travel data; The driving condition classification module is used to obtain multiple first driving condition types based on the historical travel data and the first weight, and to obtain multiple second driving condition types based on the current travel data and the second weight. The multiple first driving condition types include highway smooth driving, highway aggressive driving, urban smooth driving, and urban aggressive driving. The multiple second driving condition types include highway smooth driving, highway aggressive driving, urban smooth driving, and urban aggressive driving. The future working condition type determination module is used to determine the corresponding working condition type as a high-probability working condition type if the probability of occurrence of a certain working condition type among the plurality of first working condition types is greater than a preset high-probability threshold, and to determine the second working condition type among the plurality of second working condition types that corresponds to the high-probability working condition type as the future working condition type. The driving range estimation module is used to input the current trip data and future operating condition type into a pre-trained driving range estimation model to obtain the driving range estimation result of the electric vehicle for the current trip.

8. An electronic device, characterized in that, The electronic device includes: One or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the electric vehicle range estimation method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by the computer's processor, causes the computer to perform the electric vehicle range estimation method as described in any one of claims 1 to 6.

Images