Battery management method and device, storage medium and terminal

Abstract

The invention discloses a battery management method and device, a storage medium and a terminal. Acquiring feature data of the vehicle on the current sub-path based on the energy consumption correction information of the vehicle on the previous sub-path and the driving information of the vehicle on the current sub-path; inputting the feature data into a battery prediction model, and obtaining initial prediction energy consumption, output by the battery prediction model, of the vehicle in the current sub-path; acquiring the actual energy consumption of the vehicle in the current sub-path; and on the basis of the feature data, the initial predicted energy consumption and the actual energy consumption, updating prediction error probability distribution of the battery prediction model, and correcting the initial predicted energy consumption according to the updated prediction error probability distribution to obtain energy consumption correction information of the vehicle in the current sub-path. After one sub-path is completed, the actual energy consumption is used for correcting the model, the corrected model is used for predicting the energy consumption of the next sub-path, the model can be dynamically adjusted in combination with actual conditions, the prediction accuracy is improved, and the energy distribution and management strategy of the battery is more accurately adjusted.

Description

Technical Field

[0001] The present application relates to the technical field of new energy vehicles, and particularly to a battery management method, device, storage medium, and terminal. Background Art

[0002] With the rapid development of the market, the Battery Management System (BMS), as a core component to ensure the safe and reliable operation of electric vehicles, has become increasingly important. The battery management system realizes precise control of battery charging and discharging by real-time monitoring and managing various state parameters of the battery, such as voltage, current, temperature, etc., to avoid adverse situations such as overcharging and over-discharging, thereby optimizing the service life of the battery and ensuring the driving safety of electric vehicles. However, the existing battery management systems still face problems such as insufficient accuracy of power prediction and inaccurate energy management strategies in practical applications. Summary of the Invention

[0003] The present application provides a battery management method, device, storage medium, and terminal to solve the technical problems such as insufficient accuracy of power prediction and inaccurate energy management strategies existing in the above battery management system.

[0004] In a first aspect, an embodiment of the present application provides a battery management method, which includes:

[0005] Obtaining characteristic data of the vehicle on the current sub-path based on the energy consumption correction information of the vehicle on the previous sub-path and the driving information of the vehicle on the current sub-path;

[0006] Inputting the characteristic data into a battery prediction model to obtain the initial predicted energy consumption of the vehicle on the current sub-path output by the battery prediction model;

[0007] Obtaining the actual energy consumption of the vehicle on the current sub-path;

[0008] Updating the prediction error probability distribution of the battery prediction model based on the characteristic data, the initial predicted energy consumption, and the actual energy consumption, and correcting the initial predicted energy consumption according to the updated prediction error probability distribution to obtain the energy consumption correction information of the vehicle on the current sub-path.

[0009] In a second aspect, an embodiment of the present application provides a battery management device, which includes:

[0010] A characteristic data acquisition module, configured to obtain characteristic data of the vehicle on the current sub-path based on the energy consumption correction information of the vehicle on the previous sub-path and the driving information of the vehicle on the current sub-path;

[0011] A model prediction module, configured to input the characteristic data into a battery prediction model to obtain the initial predicted energy consumption of the vehicle on the current sub-path output by the battery prediction model;

[0012] An actual energy consumption acquisition module, configured to acquire the actual energy consumption of the vehicle on the current sub-path;

[0013] An error correction module, configured to update the predicted error probability distribution of the battery prediction model based on the feature data, the initial predicted energy consumption, and the actual energy consumption, and correct the initial predicted energy consumption according to the updated predicted error probability distribution to obtain the energy consumption correction information of the vehicle on the current sub-path.

[0014] In a third aspect, an embodiment of the present application provides a computer storage medium, where the computer storage medium stores multiple instructions, and the instructions are adapted to be loaded and executed by a processor to perform the steps of the above method.

[0015] In a fourth aspect, an embodiment of the present application provides a terminal, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the computer program is adapted to be loaded and executed by the processor to perform the steps of the above method.

[0016] The beneficial effects brought by the technical solutions provided by some embodiments of the present application at least include:

[0017] The present application provides a battery management method, which obtains the characteristic data of the vehicle on the current sub-path based on the energy consumption correction information of the vehicle on the previous sub-path and the driving information of the vehicle on the current sub-path; inputs the characteristic data into a battery prediction model to obtain the initial predicted energy consumption of the vehicle on the current sub-path output by the battery prediction model; obtains the actual energy consumption of the vehicle on the current sub-path; based on the characteristic data, the initial predicted energy consumption, and the actual energy consumption, updates the prediction error probability distribution of the battery prediction model, and corrects the initial predicted energy consumption according to the updated prediction error probability distribution to obtain the energy consumption correction information of the vehicle on the current sub-path. First, by comprehensively considering the energy consumption correction information of the vehicle on the previous sub-path and the real-time driving information of the current sub-path, the characteristic data of the current sub-path is obtained, and these characteristic data can more accurately reflect the influence of historical energy consumption data and actual driving conditions on energy consumption prediction, so as to provide more accurate data input for energy consumption prediction of the current sub-path; then, the pre-trained battery prediction model is used to process the generated characteristic data and output the initial predicted energy consumption of the vehicle on the current sub-path, which provides the initial energy consumption prediction result based on the existing battery prediction model as the basis for subsequent error correction; next, the actual energy consumption data of the vehicle on the current sub-path is actually measured and recorded, and the prediction error probability distribution of the battery prediction model is updated based on the characteristic data, the initial predicted energy consumption, and the actual energy consumption, realizing the dynamic correction of the model prediction error, which helps to continuously optimize the prediction performance of the model; further, the updated error distribution is used to correct the initial predicted energy consumption of the current sub-path, so as to obtain more accurate energy consumption correction information and improve the accuracy of energy consumption prediction of the subsequent sub-path. Through the method of the present application, after the vehicle passes through a sub-path, the model can be corrected using the actual energy consumption data, and then the corrected model is used to predict the energy consumption of the next sub-path, thereby improving the prediction accuracy of the model for battery energy consumption, so that the battery management system can more accurately manage the battery energy of the vehicle, dynamically adjust the energy distribution and management strategy of the vehicle under different driving conditions, and extend the cruising range of the vehicle. Description of the Drawings

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required to be used in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present application. For those skilled in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0019] Figure 1 It is an exemplary system architecture diagram of a battery management method provided by an embodiment of the present application;

[0020] Figure 2Schematic flowchart of a battery management method provided by an embodiment of the present application;

[0021] Figure 3 Schematic flowchart of a battery management method provided by an embodiment of the present application;

[0022] Figure 4 Overall schematic flowchart of the implementation of a battery management method provided by an embodiment of the present application;

[0023] Figure 5 Schematic flowchart of a battery management method provided by an embodiment of the present application;

[0024] Figure 6 Schematic flowchart of a model training method for a battery prediction model provided by an embodiment of the present application;

[0025] Figure 7 Schematic flowchart of a hyperparameter adjustment method for a battery prediction model provided by an embodiment of the present application;

[0026] Figure 8 Block diagram of a battery management device provided by an embodiment of the present application;

[0027] Figure 9 Schematic diagram of the structure of a terminal provided by an embodiment of the present application. Detailed implementation

[0028] To make the features and advantages of the present application more obvious and understandable, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present application.

[0029] When the following description relates to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all the implementation manners consistent with the present application. On the contrary, they are only examples of the devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0030] With the rapid development of the electric vehicle market, the battery management system can achieve charge and discharge control of the battery by monitoring and managing the battery state, optimizing the service life and performance of the battery. The battery charge prediction method in the battery management system can provide the driver with an estimate of the remaining charge and the available driving range based on factors such as the current battery state and driving habits, helping the driver better understand the vehicle's battery life, so as to charge in time when needed and avoid breakdowns caused by running out of power. At the same time, the energy management strategy can monitor and analyze the state of the electric vehicle's power system in real time, including multiple aspects such as battery charge, motor efficiency, and vehicle speed, and optimize the energy distribution through intelligent algorithms, enabling the electric vehicle to operate efficiently under different working conditions, thereby reducing energy waste and enhancing the battery life of the electric vehicle.

[0031] However, in practical applications, the battery management system still faces some technical problems, which limit the further improvement of its performance and the widespread application of electric vehicles. First, the battery state may be affected by changes in real-time traffic information such as road speed limits and lane information. Traditional battery charge prediction methods mainly rely on historical data and simple physical models for prediction, lacking the effective utilization of real-time traffic information. This makes it difficult for charge prediction to adapt to complex traffic environments and changing driving conditions, resulting in insufficient prediction accuracy. At the same time, in urban traffic, factors such as traffic congestion and waiting at traffic lights can cause a large deviation between the actual energy consumption and the predicted value. Most existing energy management strategies are designed based on standard working conditions or fixed rules, lacking the ability to dynamically adapt to actual driving conditions, resulting in the inability of existing energy management strategies to be dynamically adjusted according to traffic conditions, thus affecting the battery life of electric vehicles.

[0032] Therefore, the embodiments of this application provide a battery management method to improve the accuracy of battery charge prediction while adjusting the energy management strategy in combination with real-time traffic information, enhancing the battery life of electric vehicles.

[0033] Please refer to Figure 1 , Figure 1 which is an exemplary system architecture diagram of a battery management method provided by the embodiments of this application.

[0034] As Figure 1 shown, the system architecture may include a terminal 101, a network 102, and a server 103. The network 102 is used to provide a medium for the communication link between the terminal 101 and the server 103. The network 102 may include various types of wired communication links or wireless communication links. For example, wired communication links include optical fibers, twisted pairs, or coaxial cables, and wireless communication links include Bluetooth communication links, Wireless-Fidelity (Wi-Fi) communication links, or microwave communication links, etc.

[0035] The terminal 101 can interact with the server 103 through the network 102 to receive messages from the server 103 or send messages to the server 103, or the terminal 101 can interact with the server 103 through the network 102 to further receive messages or data sent by other users to the server 103. Exemplarily, the terminal 101 can generate the characteristic data of the current sub-path based on the energy consumption correction information of the vehicle on the previous sub-path and the driving information of the current sub-path, and send the characteristic data to the server 103 through the network 102; when the server 103 receives the characteristic data sent by the terminal 101, it will input it into the pre-trained battery prediction model to calculate the initial predicted energy consumption of the vehicle on the current sub-path, and then the server 103 will return the initial predicted energy consumption to the terminal 101 through the network 102.

[0036] The terminal 101 can be hardware or software. When the terminal 101 is hardware, it can be various electronic devices, including but not limited to in-vehicle computers, intelligent instrument panels, battery management system hardware, etc. When the terminal 101 is software, it can be installed in the above-listed electronic devices, which can be implemented as multiple software or software modules (for example, used to provide distributed services), or can be implemented as a single software or software module, and no specific limitation is made here.

[0037] In the embodiment of the present application, the terminal 101 first obtains the characteristic data of the vehicle on the current sub-path based on the energy consumption correction information of the vehicle on the previous sub-path and the driving information of the vehicle on the current sub-path; then, the terminal 101 inputs the characteristic data into the battery prediction model to obtain the initial predicted energy consumption of the vehicle on the current sub-path output by the battery prediction model; next, the terminal 101 will obtain the actual energy consumption of the vehicle on the current sub-path; finally, the terminal 101 updates the prediction error probability distribution of the battery prediction model based on the characteristic data, the initial predicted energy consumption, and the actual energy consumption, and corrects the initial predicted energy consumption according to the updated prediction error probability distribution to obtain the energy consumption correction information of the vehicle on the current sub-path.

[0038] The server 103 can be a business server that provides various services. It should be noted that the server 103 can be hardware or software. When the server 103 is hardware, it can be implemented as a distributed server cluster composed of multiple servers, or can be implemented as a single server. When the server 103 is software, it can be implemented as multiple software or software modules (for example, used to provide distributed services), or can be implemented as a single software or software module, and no specific limitation is made here.

[0039] Alternatively, the system architecture may not include the server 103. In other words, the server 103 may be an optional device in the embodiments of this specification. That is, the method provided in the embodiments of this specification may be applied to a system structure that only includes the terminal 101, and the embodiments of this application do not make any limitations in this regard.

[0040] It should be understood that Figure 1 the number of terminals, networks, and servers in

[0041] Please refer to Figure 2 , Figure 2 which is a schematic flowchart of a battery management method provided by the embodiments of this application. The execution subject of the embodiments of this application may be a terminal that executes battery management, or a processor in the terminal that executes the battery management method, or a battery management service in the terminal that executes the battery management method. For the convenience of description, hereinafter, taking the execution subject as the processor in the terminal as an example, the specific execution process of the battery management method will be introduced.

[0042] As Figure 2 shown, the battery management method may at least include:

[0043] S202. Obtain the characteristic data of the vehicle on the current sub-path based on the energy consumption correction information of the vehicle on the previous sub-path and the driving information of the vehicle on the current sub-path.

[0044] Optionally, since there may be different traffic conditions, road conditions (such as speed limits, curvatures, etc.) in different stages of the global driving path, these factors will affect the actual energy consumption of the vehicle battery. Therefore, during the actual driving of the vehicle, the initial predicted energy consumption reflecting the change of the vehicle battery state may have deviations. The method of the embodiments of this application divides the global path of the vehicle into multiple sub-paths, and conducts more detailed energy consumption analysis on the specific conditions of each sub-path, and then corrects according to the actual energy consumption after the end of each sub-path, so as to reduce the cumulative error and make the energy consumption prediction of the subsequent sub-paths closer to the actual situation.

[0045] Specifically, the actual driving information of the current sub-path, such as road conditions, traffic flow, battery conditions, etc., directly affects the energy consumption of the vehicle battery. For example, frequent starts and stops in congested sections will result in higher energy consumption, while driving at a constant speed on the highway is relatively energy-saving. Therefore, collecting and analyzing the actual driving information of the current sub-path in real time and extracting features in combination with these actual driving information can make the energy consumption prediction closer to the actual driving situation.

[0046] Optionally, the energy consumption of the vehicle battery is not only affected by the current actual driving information, but also closely related to its previous operating state. By continuously updating and learning the energy consumption correction information of the previous path (i.e., the previous sub-path), the changing trend of the battery performance can be better captured and the error can be corrected in real time. For example, due to reasons such as battery aging and environmental temperature changes, it is found that the actual energy consumption of the battery in the previous sub-path is higher than the predicted value. Incorporating this information in the previous sub-path into the feature data of the current sub-path helps to more accurately adjust the energy consumption prediction of the current sub-path. If the energy consumption correction information of the previous sub-path is ignored, the battery prediction model will not be able to self-adjust based on previous experience, which may lead to long-term cumulative prediction biases.

[0047] Optionally, when predicting the energy consumption of the current sub-path where the vehicle is located, first obtain the energy consumption correction information of the vehicle during the driving process of the previous sub-path, and collect and record the driving information of the vehicle in the current sub-path in real time through the sensors built in the vehicle or the Global Positioning System (GPS). This includes but is not limited to battery state data such as the state of charge and charge-discharge power of the vehicle battery, and road speed limits, lane information, curvature, etc. of the current sub-path obtained through the map navigation software. Then fuse the above energy consumption correction information and driving information, and perform preliminary processing through a preset data processing algorithm (such as normalization, filtering, etc.) to remove noise and highlight key features, so as to construct a high-dimensional feature data vector, which provides more accurate and comprehensive input information for the subsequent battery prediction model.

[0048] It should be noted that the energy consumption correction information before the current sub-path can be the energy consumption correction information of the previous sub-path, or the energy consumption correction information of several or all sub-paths before the current sub-path. Which stages of energy consumption correction information to specifically select is flexibly adjusted according to the actual situation or prediction accuracy, and the embodiments of the present application do not limit this.

[0049] S204. Input the feature data into the battery prediction model to obtain the initial predicted energy consumption of the vehicle in the current sub-path output by the battery prediction model.

[0050] Optionally, considering that the battery prediction model may generate errors, in order to correct these errors in real time, the initial predicted energy consumption of the current sub-path predicted by the model and its corresponding actual energy consumption can be collected and recorded respectively. Then, by comparing the actual energy consumption with the initial predicted energy consumption, the prediction error of the battery prediction model can be reflected, and then the model can be corrected through this error information to reduce subsequent prediction errors.

[0051] Specifically, after obtaining the feature data of the vehicle on the current sub-path, this feature data is used as an input variable and fed into a pre-trained battery prediction model. After receiving the data, the battery prediction model will calculate and output the initial predicted energy consumption of the vehicle on the current sub-path based on the rules or patterns learned internally. This predicted value reflects the energy level that the vehicle's battery is expected to consume under given conditions and provides a basic reference value for subsequent correction of the energy consumption prediction error.

[0052] S206. Obtain the actual energy consumption of the vehicle on the current sub-path.

[0053] Optionally, the energy consumption of the vehicle on the current sub-path is monitored in real time through on-vehicle sensors (such as current sensors, voltage sensors, etc.), such as the charge and discharge state of the battery, the working efficiency of the motor, and other relevant energy consumption parameters. After the vehicle has completed the entire current sub-path, the actual energy consumption of the vehicle in the entire current sub-path is calculated based on the data monitored in real time. This actual energy consumption value is used to verify the accuracy of the initial predicted energy consumption and thus evaluate the performance of the battery prediction model.

[0054] S208. Update the probability distribution of the prediction error of the battery prediction model based on the feature data, the initial predicted energy consumption, and the actual energy consumption, and correct the initial predicted energy consumption according to the updated probability distribution of the prediction error to obtain the energy consumption correction information of the vehicle on the current sub-path.

[0055] Optionally, since it is necessary to correct the energy consumption prediction error of the model, the method in the embodiments of the present application will construct a probability distribution of the prediction error based on the prediction error records in the historical data to describe the uncertainty of the prediction error. After the actual energy consumption data of the current sub-path is collected, this probability distribution can be updated according to the current prediction error through probability statistical methods such as Bayesian update, so that the distribution of the prediction error is closer to the actual situation of the current sub-path to reflect the prediction possibility of the battery prediction model under the current feature data conditions.

[0056] Furthermore, based on the updated probability distribution of the prediction error, the initial predicted energy consumption value of the current sub-path is adjusted to obtain the energy consumption correction information of the vehicle in the current sub-path, which can be achieved by calculating the expected value of the prediction error or generating a more reliable prediction interval according to the probability distribution of the prediction error. For example, if the updated probability distribution of the prediction error indicates that the initial predicted energy consumption output by the model is generally higher than the actual energy consumption, the initial predicted energy consumption can be appropriately lowered to compensate for this deviation. At the same time, the corrected energy consumption data corresponding to the current sub-path can also be used as reference information for the prediction of the next sub-path and fed back to the battery prediction model to form a closed-loop optimization mechanism.

[0057] In an embodiment of the present application, a battery management method is provided. Feature data of the vehicle on the current sub-path is obtained based on the energy consumption correction information of the vehicle on the previous sub-path and the driving information of the vehicle on the current sub-path; the feature data is input into a battery prediction model to obtain the initial predicted energy consumption of the vehicle on the current sub-path output by the battery prediction model; the actual energy consumption of the vehicle on the current sub-path is obtained; based on the feature data, the initial predicted energy consumption, and the actual energy consumption, the prediction error probability distribution of the battery prediction model is updated, and the initial predicted energy consumption is corrected according to the updated prediction error probability distribution to obtain the energy consumption correction information of the vehicle on the current sub-path. First, by comprehensively considering the energy consumption correction information of the vehicle on the previous sub-path and the real-time driving information of the current sub-path, the feature data of the current sub-path is obtained, and these feature data can more accurately reflect the influence of historical energy consumption data and actual driving conditions on energy consumption prediction, so as to provide more accurate data input for the energy consumption prediction of the current sub-path; then, the pre-trained battery prediction model is used to process the generated feature data and output the initial predicted energy consumption of the vehicle on the current sub-path, which provides the initial energy consumption prediction result based on the existing battery prediction model as the basis for subsequent error correction; next, the actual energy consumption data of the vehicle on the current sub-path is actually measured and recorded, and the prediction error probability distribution of the battery prediction model is updated based on the feature data, the initial predicted energy consumption, and the actual energy consumption, realizing the dynamic correction of the model prediction error, which helps to continuously optimize the prediction performance of the model; further, the updated error distribution is used to correct the initial predicted energy consumption of the current sub-path, so as to obtain more accurate energy consumption correction information and improve the accuracy of the energy consumption prediction of the subsequent sub-path. Through the method of the present application, after the vehicle passes through a sub-path, the model can be corrected using the actual energy consumption data, and then the corrected model is used to predict the energy consumption of the next sub-path, so as to improve the prediction accuracy of the model for battery energy consumption. In this way, the battery management system can more accurately manage the battery energy of the vehicle to dynamically adjust the energy distribution and management strategy of the vehicle under different driving conditions and extend the cruising range of the vehicle.

[0058] Please refer to Figure 3 , Figure 3 which is a schematic flowchart of a battery management method provided by an embodiment of the present application.

[0059] As Figure 3 shown, the battery management method may at least include:

[0060] S302. Obtain the complete path of the vehicle through a map navigation software, and divide the complete path into multiple sub-paths according to traffic nodes or a preset length.

[0061] Optionally, Figure 4It is a schematic diagram of the overall process implemented by a battery management method provided by an embodiment of the present application. Among them, S404 - S406 is the process of training a neural network model based on human driving historical data, and S408 - S412 is the process of predicting energy consumption during the actual operation of the vehicle based on a pre-trained model. As Figure 4 As shown in S408 in, the method in the embodiment of the present application calls mainstream map navigation software on the market (such as Amap, Baidu Map, etc.) to obtain the complete path information of the vehicle from the starting position to the destination according to a preset data collection time (for example, 10 seconds), which at least includes the following variables: distance Dis, estimated travel time Time, road speed limit Speed, lane information L_info (such as the number of lanes and layout information, etc.), traffic rules Tra (such as no left turn, no right turn, etc.), curvature Cur, real-time traffic flow density Tra_flow, real-time signal light status Tra_t (such as red light, green light, etc.).

[0062] Further, in order to improve the accuracy of energy consumption estimation, the complete path is divided into multiple sub-path segments according to traffic nodes (such as traffic lights, intersections) or a preset length (for example, per kilometer). Exemplarily, for urban roads or structured roads, they are divided according to key nodes such as traffic lights and intersections; while for general roads, each kilometer is divided into a sub-path. Among them, each sub-path contains detailed feature information, and the specific content is the same as the variable content included in the aforementioned complete path information.

[0063] S304. Obtain the traffic characteristics of the current sub-path and the battery characteristics of the vehicle on the current sub-path, and determine the feature data of the vehicle on the current sub-path according to the energy consumption correction information, traffic characteristics, and battery characteristics of the vehicle on the previous sub-path.

[0064] Optionally, in addition to real-time collection of the feature information (traffic characteristics) of each sub-path, the method in the embodiment of the present application can also real-time collect (for example, with a data collection time interval of 10 seconds) the battery state data (battery characteristics) of the vehicle through in-vehicle sensors, which at least includes information such as voltage V, current I, temperature T, state of charge SOC, charge and discharge power Pb, etc. Based on this, when predicting the energy consumption of the current sub-path, first obtain the traffic characteristics of the current sub-path and the battery characteristics of the vehicle on the current sub-path, and then extract effective features that can be used for subsequent analysis through data preprocessing steps.

[0065] Specifically, the data preprocessing steps may include mean processing, removing outliers, etc. Among them, assume that the unknown variable x is used to represent the traffic characteristics and battery characteristics of the currently collected sub-path, and these data are processed using a mean filter. The mean filter smooths the data by calculating the average value of consecutive N data points, and the calculation formula is as follows: where N is the window size of the mean filter, and x t-1 to x t-N are the data from time t - 1 to t - N. This step can effectively remove some noise in the data, making the data better reflect the real state during vehicle driving. At the same time, during the actual operation of the vehicle, the decay change of the state of charge (SOC) is slow. Therefore, when the SOC satisfies |Δ| < 0.05% (0.05% is the change threshold of the SOC during data acquisition), the corresponding data is retained; otherwise, the data is removed. And for all the collected parameter data, outliers are identified and removed according to the relationship between the data and the mean and standard deviation. For example, for the data x to be processed, if it satisfies where x mean is the data mean, and x std is the data standard deviation, then it is determined as an outlier and removed to ensure the reliability of the data.

[0066] Optionally, after obtaining the traffic features and battery features corresponding to the current sub - path, the feature data set of the current sub - path can be determined by combining the energy consumption correction information of the vehicle on the previous sub - path. Exemplarily, all the above - mentioned features are fused to form a high - dimensional feature vector x (feature data): where V is the current voltage value of the battery; I is the current current value of the battery; T is the current temperature value of the battery; SOC is the current state of charge of the battery; Pb is the current charge - discharge power of the battery; Dis is the distance of the sub - path; Time is the estimated passing time of the sub - path; Speed is the road speed limit of the current section; L_info is the vector representation of the number of lanes and layout information of the current section; Tra is the vector representation of the traffic rules of the current section, such as no left - turn, no right - turn, etc.; Cur is the curvature information of the current section; Tra_flow is the real - time traffic flow density of the current section; Tra_t is the real - time signal light state time of the current section; is the energy consumption estimation result (corrected predicted energy consumption) of the previous sub - path, reflecting the current energy consumption state; α is the error correction coefficient, which is dynamically updated based on Bayesian estimation to adjust the accuracy of energy consumption estimation.

[0067] S306. Input the feature data into the battery prediction model to obtain the initial predicted energy consumption of the vehicle on the current sub - path output by the battery prediction model; obtain the actual energy consumption of the vehicle on the current sub - path.

[0068] Optionally, after obtaining the feature data of the vehicle on the current sub - path, this feature data is used as an input variable and sent into a pre - trained battery prediction model. After receiving the data, the battery prediction model will calculate and output the initial predicted energy consumption of the vehicle on the current sub - path based on the rules or patterns learned internally. Further, asFigure 4 As shown in S410, after the vehicle has completed the entire current sub-path, the actual energy consumption data of the vehicle in the entire current sub-path is collected to verify the accuracy of the initial predicted energy consumption, thereby evaluating the performance of the battery prediction model.

[0069] Exemplarily, assume that the vehicle starts from point A, and the global path is divided into n sub-paths, denoted as S1, S2,..., S n . Assume that the current sub-path is the i-th sub-path S i , and the initial predicted energy consumption output by the battery prediction model is The calculation formula is as follows: Where, is the predicted value of the vehicle's energy consumption by the battery prediction model in the current sub-path (initial predicted energy consumption), x i is the characteristic data of the current sub-path, specifically including the parameters as described in S304, and θ is the model hyperparameter obtained after the heuristic algorithm search. Further, after the vehicle passes through the current sub-path, the actual energy consumption of the vehicle in the entire current sub-path is collected: y i， .

[0070] S308. Calculate the prediction error between the actual energy consumption and the initial predicted energy consumption; based on the prediction error and the characteristic data, update the prior prediction error probability distribution of the battery prediction model to the posterior prediction error probability distribution according to Bayes' theorem.

[0071] Optionally, assume that in the previous sub-path, the prediction error probability distribution of the battery prediction model follows a certain prior distribution (for example, Gaussian distribution: ), and this prior distribution reflects the uncertainty estimation of the prediction error before the observation data of the current sub-path. In order to continuously improve the accuracy of the battery prediction model, as Figure 4 shown in S410, the method shown in the embodiments of the present application uses Bayes' theorem and combines the actual energy consumption and real-time information to update the probability distribution of the prediction error.

[0072] Still taking the example shown in S306, after obtaining the actual energy consumption of the current sub-path, further calculate the prediction error e i between the initial predicted energy consumption output by the battery prediction model and the actual energy consumption, and the calculation formula is as follows: Then use the Bayesian network to process the prediction error, and continuously update the probability distribution of the prediction error based on the historical data during driving. Assume that P(e i |x i ) is the probability distribution of the prediction error e i under the given characteristic data x i . According to Bayes' theorem, the updated probability distribution P(e i |xi ) can be calculated as follows: where P(x i |e i ) is the conditional probability of the given prediction error e i for the feature data x i , P(e i ) is the prior probability of the prediction error e i , and P(x i ) is the marginal probability of the feature data x i . Under the assumption of a Gaussian distribution, the posterior distribution P(e i |x i ) also follows a Gaussian distribution and satisfies: where e i,j is the n prediction error data newly obtained for the current sub-path, is the variance of the corresponding prediction error. The above formula reflects the update of the mean and variance by combining the new observation data of the current sub-path. As new data is added, the mean and variance will be continuously adjusted to more accurately reflect the change in the distribution probability of the prediction error.

[0073] It should be noted that since the characteristic quantities to be concerned about in the error correction stage focus more on the parameters that dynamically change during vehicle driving, the feature data used in Bayesian estimation can be as follows: voltage V, current I, state of charge SOC, estimated passing time Time of the sub-path, real-time traffic flow density Tra_flow, real-time traffic signal state Tra_t.

[0074] S310. Determine the error correction coefficient based on the expectation of the prediction error in the posterior prediction error probability distribution, and correct the initial predicted energy consumption based on the error correction coefficient to obtain the corrected predicted energy consumption; use the corrected predicted energy consumption and the error correction coefficient as the energy consumption correction information of the vehicle on the current sub-path.

[0075] Optionally, after obtaining the updated prediction error probability distribution, the initial predicted energy consumption of the battery prediction model can be corrected by combining this probability distribution, so as to obtain a more accurate corrected predicted energy consumption. Specifically, extract the expectation of the prediction error (i.e., the mean of the posterior prediction error probability distribution) from the updated posterior prediction error probability distribution. This expectation represents the best estimate of the prediction error of the model for the current sub-path and can also be used as the error correction coefficient for the current sub-path. Then, adjust the initial predicted energy consumption of the current sub-path according to this error correction coefficient to compensate for the prediction error of the battery prediction model.

[0076] Still taking the example shown in S306, after obtaining the updated prediction error probability distribution of the current sub-path, the error correction coefficient of the current sub-path can be determined as μi , at this time, the initial predicted energy consumption for the current sub-path Adjust the predicted value according to the updated predicted error probability distribution. The calculation formula is as follows: Where is the corrected predicted energy consumption.

[0077] Optionally, use the corrected predicted energy consumption of the current sub-path and the error correction coefficient as the energy consumption correction information of the vehicle on the current sub-path. These information not only provide a more accurate energy consumption estimate on the current sub-path, but also provide a reference for the energy consumption prediction of subsequent sub-paths. By continuously repeating the above online update process, the method in the embodiment of the present application can continuously optimize its prediction performance according to real-time driving data and environmental information, so as to more accurately estimate the passing energy consumption of the vehicle on different sections and provide a more reliable basis for the battery prediction of electric vehicles.

[0078] S312. Obtain the predicted state of charge output by the battery prediction model, and adjust the power of the battery on the corresponding sub-path based on the predicted state of charge and the traffic characteristics on the corresponding sub-path.

[0079] Optionally, as Figure 4 shown in S412, in addition to obtaining the initial predicted energy consumption, the battery prediction model also outputs the predicted state of charge of the vehicle on the current sub-path This predicted value reflects the percentage of the remaining battery power expected under the current conditions of the vehicle. Based on this, the charging and discharging power of the battery can be dynamically adjusted according to the predicted state of charge and the traffic characteristics corresponding to the current sub-path to optimize the energy use efficiency and increase the driving range of the vehicle. Exemplarily, the method in the embodiment of the present application takes four situations of traffic lights intersections, roads with large curvature, crowded sections and long downhill roads as examples to illustrate how to adjust the output power of the battery in combination with the traffic characteristics corresponding to the current sub-path.

[0080] Optionally, under traffic congestion conditions, the vehicle starts and stops frequently. Based on the real-time traffic characteristics provided by the map, the battery needs to maintain a medium and low power output to prevent the vehicle from starting too violently. Obtain the real-time traffic flow density Tra_flow of the current section from the map navigation software, set a traffic flow density threshold, and when the real-time traffic flow density approaches the maximum traffic flow density threshold, the battery output power gradually decreases to prevent the vehicle from starting too violently. The calculation formula is as follows: Where, P b is the current battery output power, P max is the maximum output power of the battery, Tra_flow max is the preset maximum traffic flow density threshold.

[0081] Optionally, on a road with a large curvature turn where the vehicle is about to make a turn without a view, the battery output power should be appropriately reduced before entering the curve to reduce the power supply of the vehicle. As the vehicle gets closer to the curve entrance, the battery output power gradually decreases. Eventually, when approaching the curve entrance, the power drops to a certain lower limit determined by the curvature. This pre-limitation of power can prevent the driver from suddenly braking at the curve turning point. The calculation formula is as follows: Where, P b is the current battery output power, P max is the maximum output power of the battery, Cur max is the preset maximum curvature threshold, Dis is the distance between the vehicle and the curve entrance, and there is a set lower limit.

[0082] Optionally, at a traffic light intersection, when the light is red and the waiting time is too long, the battery power is limited before the vehicle reaches the intersection to reduce unnecessary energy consumption. Therefore, the power limitation strategy in this scenario mainly considers the remaining waiting time of the traffic light. The calculation formula is as follows: Where, P b is the current battery output power, P max is the maximum output power of the battery, tra_t max is the maximum waiting time for the red light at this location, and tra_t is the remaining waiting time at the current intersection.

[0083] Optionally, on sections such as long downhill roads, a more aggressive regenerative braking strategy can be adopted to limit the vehicle speed while charging. A fixed regenerative braking coefficient K is used during normal driving normal to recover energy as much as possible. On long downhill sections, the regenerative braking coefficient is determined by the slope and the length of the downhill section together. The calculation formula is as follows: Where, α and β are weighting coefficients used to balance the influence of the slope and the length of the downhill section, Slope max is the preset maximum slope threshold, Dis is the length of the downhill section in the current sub-section statistically obtained based on path splitting, and Len represents the total length of the sub-section.

[0084] In an embodiment of the present application, a battery management method is provided. By obtaining the complete path of the vehicle through a map navigation software and dividing the path into multiple sub-paths according to traffic nodes or a preset length, the specific conditions of different road sections can be captured more precisely, providing more accurate data support for subsequent energy management. At the same time, by combining the traffic characteristics and battery characteristics of the current sub-path, the energy consumption characteristics under actual driving conditions can be better reflected, enabling the battery prediction model to dynamically respond to external changes, adapt to various complex road conditions and environmental conditions, and improve the accuracy of energy consumption prediction. Further, by calculating the prediction error between the actual energy consumption and the initial predicted energy consumption and applying Bayes' theorem to update the probability distribution of the prediction error, the model can be dynamically adjusted in combination with the actual situation to make the prediction result closer to the actual situation. Finally, by combining the predicted state of charge and traffic characteristics, the charging and discharging power of the battery is dynamically adjusted. This precise energy management and power regulation can optimize the energy usage efficiency and help extend the vehicle's cruising range.

[0085] Please refer to Figure 5 , Figure 5 which is a schematic flow chart of a battery management method provided by an embodiment of the present application;

[0086] As Figure 5 shown, the battery management method may at least include:

[0087] S502. Construct an initial battery prediction model based on a long short-term memory network and a bidirectional gated recurrent unit.

[0088] Optionally, the method of the embodiment of the present application adopts a neural network model combining a long short-term memory network (LSTM) and a bidirectional gated recurrent unit (BiGRU) as the battery prediction model to predict the change in the state of charge of the battery. Based on this, first construct a neural network model combining LSTM and BiGRU, where the LSTM layer is responsible for capturing the long-term dependencies of time series data, and the BiGRU layer is responsible for combining historical and future information to improve the accuracy of time series modeling. Figure 6 which is a schematic flow chart of a model training method for a battery prediction model provided by an embodiment of the present application, where S610 is an example of the structure of the initial battery prediction model.

[0089] Specifically, traditional Recurrent Neural Networks (RNNs) have internal states or memories. At each time step, an RNN not only considers the current input but also the information from previous time steps, enabling them to process input sequences with temporal dependencies, such as natural language, time series data, etc. Although RNNs can handle sequence data, they have limitations in dealing with long sequences, such as the vanishing gradient problem, which leads to the inability of RNNs to effectively capture long-term dependencies in sequences. LSTM is a special type of recurrent neural network that can effectively solve the vanishing gradient problem of RNNs by introducing memory cells and gating mechanisms to capture long-term dependencies in time series data. During vehicle driving, the change in the state of charge of the battery is a typical time series problem affected by various factors such as historical driving data, driving habits, and traffic conditions. LSTM can effectively handle such long-term dependencies by introducing memory cells and gating mechanisms.

[0090] Optionally, LSTM mainly consists of four parts: a forget gate, an input gate, memory cell update, and an output gate. The forget gate determines what information to discard from the memory cell, and its calculation formula is as follows: f t =σ(W f ·[h t-1 ,x t +b f ), where f t is the output of the forget gate, W f and b f are the weight matrix and bias term respectively, h t-1 is the hidden state at the previous moment, and x t is the input at the current moment. The input gate determines which new information will be stored in the memory cell, and its calculation formula is as follows: i t =σ(W i ·[h t-1 ,x t +b i ), where i t is the output of the input gate, is the value of the candidate memory cell, and W i , b i , W c , b c are the weight matrix and bias term respectively. Memory cell update is used to update the state of the memory cell, and its calculation formula is as follows: where c t is the memory cell at the current moment, and ⊙ represents element-wise multiplication. The output gate calculates the output at the current time step based on the state of the memory cell, and its calculation formula is as follows: o t= σ(W o · [h t-1 , x t + b o ), h t = o t ⊙ tanh(c t ), where o t is the output of the output gate, h t is the hidden state at the current time step, W o and b o are the weight matrix and bias term respectively.

[0091] Optionally, BiGRU is an improved RNN that can capture information in time series data in both forward and backward directions by introducing a gating mechanism, thus solving the vanishing gradient problem and improving the performance of the model by transmitting information bidirectionally. The reset gate determines how much past information to ignore, and its calculation formula is as follows: r t = σ(W r · [h t-1 , x t + b r ), where r t is the output of the reset gate, W r and b r are the weight matrix and bias term respectively. The update gate determines how much past hidden state to retain, and its calculation formula is as follows: z t = σ(W z · [h t-1 , x t + b z ), where z t is the output of the update gate, W z and b z are the weight matrix and bias term respectively. The candidate hidden state determines what the new hidden state should be if part of the past information is ignored, and its calculation formula is as follows: where, is the candidate hidden state, W h and b h are the weight matrix and bias term respectively. The hidden state update determines how much past information to retain and how much new candidate hidden state to combine to form the final hidden state at the current time step, and its calculation formula is as follows: where, h t is the hidden state at the current time step. The BiGRU layer contains GRU units in two directions, processing time series data from the forward and backward directions respectively. The output of the forward GRU unit is denoted as The output of the backward GRU unit is denoted as The final hidden state h t is the concatenation of the forward and backward hidden states:

[0092] S504. Obtain multiple sample feature data, and input each sample feature data into the initial battery prediction model to train the initial battery prediction model.

[0093] Optionally, Figure 6 As shown in S602 in Figure 6 , before training the constructed initial battery prediction model in the embodiments of the present application, it is first necessary to collect a large amount of human driving data, fuse these data into sample feature data, and then input these sample feature data into the initial battery prediction model for training. Specifically, regarding how to collect data and fuse it into sample feature data, please refer to the detailed description in steps S302 and S304, which will not be elaborated here.

[0094] S506. During the training process of the initial battery prediction model, control the initial battery prediction model to output the corresponding predicted state of charge based on each sample feature data, and adjust the hyperparameters of the initial battery prediction model according to each predicted state of charge and the true state of charge of each sample feature data until the initial battery prediction model converges, and obtain the trained battery prediction model.

[0095] Optionally, as shown in S604 - S614 in Figure 6 , during the training process, the sample feature data is used as the input of the initial battery prediction model, and the model outputs the predicted state of charge of the battery Then, by comparing the predicted state of charge with its corresponding true state of charge, the hyperparameters of the initial battery prediction model are continuously adjusted to obtain the trained battery prediction model. To ensure the training effect of the model, the historical data and the working condition prediction data are divided into a training set and a test set. The training set data is used to train the initial battery prediction model, and the test set data is used to verify the performance of the model.

[0096] Optionally, during the training process of the battery prediction model, the error of the model can be evaluated by defining a loss function. Specifically, for each sample feature data, the data is input into the initial battery prediction model to obtain the corresponding predicted state of charge, and then the training loss value is calculated by comparing the predicted state of charge with the true state of charge. The loss function used can be the mean squared error (MSE), and the calculation formula is as follows: where N is the number of samples, SOC i is the true state of charge, is the predicted state of charge output by the model.

[0097] Optionally, Figure 7The flowchart of a method for adjusting hyperparameters of a battery prediction model provided by an embodiment of the present application is shown as Figure 7 shown in S702 - S712 in

[0098] Exemplarily, as Figure 7 shown in S706 in

[0099] Specifically, for the established initial battery prediction model, first randomly generate a set of initial moth positions (X i )(i = 1, 2, …, n), where n is the number of moths. Each moth represents a potential hyperparameter combination. For example, the hyperparameter combination can include ranges such as the learning rate η, batch size B, the number of neurons in the hidden layer H, etc. Then for each set of hyperparameter combinations, use these hyperparameters to train the initial battery prediction model, and calculate the training loss value MSE of the model on the validation set as the fitness value f(X i ). Since the fitness value is related to the performance metrics of the model in the prediction task, the lower the fitness value, the better the model prediction effect under this hyperparameter combination. Therefore, sort the moths according to the fitness value, and select the first n / 2 moths as the flames (F j )(j = 1, 2, …, n / 2), and this flame represents the current optimal solution. Next, for each moth, update its position according to the position of the flame, gradually approaching the global optimal solution. The calculation formula is as follows: D ij = |F j - X i |, X i = F j - D ij ·e -bt·cos(2πt), where b is a constant, t is the current iteration number, and D ij is the distance between the moth X i and the flame F j . Further, calculate the fitness value of the updated moth, update the position of the flame, and select the current optimal moth as the new flame. Repeat the above steps until the maximum number of iterations is reached or other termination conditions are met (such as the prediction error is less than a preset threshold). After multiple iterations, select the position of the moth with the lowest fitness value as the best hyperparameter combination. At this time, it is considered that the model has converged, and the finally trained battery prediction model is obtained.

[0100] Optionally, as shown in S616 - S622 in Figure 6 , after obtaining the trained battery prediction model, the prediction error of the model can be further evaluated through error analysis to determine whether the model needs to be further adjusted or additional training is required. And based on the results of the error analysis, adjust the hyperparameters of the model to improve the prediction performance, and optimize the model to better adapt to the data and task requirements. Then divide the dataset into a training set, a validation set, and a test set to evaluate the generalization ability of the model, ensure that the model can also perform well on unseen data, prevent the model from overfitting, and improve the robustness and reliability of the model. Next, retrain the model with the adjusted hyperparameters, input the sample feature data into the model, and perform a new round of training to update the model to reflect the new hyperparameter settings and improve the prediction accuracy.

[0101] Optionally, as shown in S714 - S716 in Figure 7 , after determining the trained battery prediction model, the energy consumption of the battery during the actual driving process of the vehicle can be predicted based on this model, and the prediction error of the model can be evaluated according to the prediction results to determine whether the model needs to be further adjusted or additional training is required.

[0102] In the embodiment of the present application, a battery management method is provided. An initial battery prediction model is constructed by combining LSTM and BiGRU. LSTM can capture the long - term trends and patterns in the battery state changes, while BiGRU can capture the information of time - series data from both forward and backward directions, thus improving the prediction accuracy of the state of charge and energy consumption of the battery; an improved moth - flame optimization algorithm and other meta - heuristic algorithms are used to efficiently search the hyperparameter space, so as to optimize the hyperparameters of the battery prediction model and improve the training efficiency and prediction accuracy of the model.

[0103] Please refer to Figure 8 , Figure 8 , which is the structural block diagram of a battery management device provided by the embodiment of the present application.

[0104] As shown in Figure 8As shown, the battery management device 800 includes:

[0105] A feature data acquisition module 810, configured to acquire the feature data of the vehicle on the current sub-path based on the energy consumption correction information of the vehicle on the previous sub-path and the driving information of the vehicle on the current sub-path;

[0106] A model prediction module 820, configured to input the feature data into the battery prediction model to obtain the initial predicted energy consumption of the vehicle on the current sub-path output by the battery prediction model;

[0107] An actual energy consumption acquisition module 830, configured to acquire the actual energy consumption of the vehicle on the current sub-path;

[0108] An error correction module 840, configured to update the predicted error probability distribution of the battery prediction model based on the feature data, the initial predicted energy consumption, and the actual energy consumption, and correct the initial predicted energy consumption according to the updated predicted error probability distribution to obtain the energy consumption correction information of the vehicle on the current sub-path.

[0109] In some possible embodiments, the battery management device 800 further includes: a path division module, configured to obtain the complete path of the vehicle through a map navigation software, and divide the complete path into multiple sub-paths according to traffic nodes or a preset length; the feature data acquisition module 810 is further configured to acquire the traffic feature of the current sub-path and the battery feature of the vehicle on the current sub-path, and determine the feature data of the vehicle on the current sub-path according to the energy consumption correction information, the traffic feature, and the battery feature of the vehicle on the previous sub-path.

[0110] In some possible embodiments, the error correction module 840 is further configured to calculate the prediction error between the actual energy consumption and the initial predicted energy consumption; based on the prediction error and the feature data, update the prior predicted error probability distribution of the battery prediction model to a posterior predicted error probability distribution according to Bayes' theorem.

[0111] In some possible embodiments, the error correction module 840 is further configured to determine an error correction coefficient according to the expectation of the prediction error in the posterior predicted error probability distribution, and correct the initial predicted energy consumption based on the error correction coefficient to obtain a corrected predicted energy consumption; use the corrected predicted energy consumption and the error correction coefficient as the energy consumption correction information of the vehicle on the current sub-path.

[0112] In some possible embodiments, the battery management device 800 further includes: a model training module, configured to build an initial battery prediction model based on a long short-term memory network and a bidirectional gated recurrent unit; obtain a plurality of sample feature data, and input each sample feature data into the initial battery prediction model to train the initial battery prediction model; during the training process of the initial battery prediction model, control the initial battery prediction model to output corresponding predicted state of charge based on each sample feature data, and adjust the hyperparameters of the initial battery prediction model until the initial battery prediction model converges according to each predicted state of charge and the true state of charge of each sample feature data, so as to obtain a trained battery prediction model.

[0113] In some possible embodiments, the model training module is further configured to calculate a training loss value according to each predicted state of charge and the true state of charge of each sample feature data, and adjust the hyperparameters of the initial battery prediction model through a metaheuristic algorithm based on the training loss value until the initial battery prediction model converges.

[0114] In some possible embodiments, the battery management device 800 further includes: a power adjustment module, configured to obtain the predicted state of charge output by the battery prediction model, and adjust the power of the battery on the corresponding sub-path based on the predicted state of charge and the traffic characteristics on the corresponding sub-path.

[0115] In an embodiment of the present application, a battery management device is provided. Among them, a feature data acquisition module is configured to acquire the feature data of the vehicle on the current sub-path based on the energy consumption correction information of the vehicle on the previous sub-path and the driving information of the vehicle on the current sub-path; a model prediction module is configured to input the feature data into a battery prediction model to obtain the initial predicted energy consumption of the vehicle on the current sub-path output by the battery prediction model; an actual energy consumption acquisition module is configured to acquire the actual energy consumption of the vehicle on the current sub-path; an error correction module is configured to update the prediction error probability distribution of the battery prediction model based on the feature data, the initial predicted energy consumption, and the actual energy consumption, and correct the initial predicted energy consumption according to the updated prediction error probability distribution to obtain the energy consumption correction information of the vehicle on the current sub-path. First, the feature data acquisition module obtains the feature data of the current sub-path by comprehensively considering the energy consumption correction information of the vehicle on the previous sub-path and the real-time driving information of the current sub-path. These feature data can more accurately reflect the influence of historical energy consumption data and actual driving conditions on energy consumption prediction, so as to provide more accurate data input for the energy consumption prediction of the current sub-path; then the model prediction module processes the generated feature data using a pre-trained battery prediction model and outputs the initial predicted energy consumption of the vehicle on the current sub-path, which provides an initial energy consumption prediction result based on the existing battery prediction model as the basis for subsequent error correction; next, the actual energy consumption acquisition module actually measures and records the actual energy consumption data of the vehicle on the current sub-path, and the error correction module updates the prediction error probability distribution of the battery prediction model based on the feature data, the initial predicted energy consumption, and the actual energy consumption, realizing dynamic correction of the model prediction error, which helps to continuously optimize the prediction performance of the model; further, the updated error distribution is used to correct the initial predicted energy consumption of the current sub-path, so as to obtain more accurate energy consumption correction information and improve the accuracy of the energy consumption prediction of the subsequent sub-path. Through the method of the present application, after the vehicle passes through a sub-path, the model can be corrected using the actual energy consumption data, and then the corrected model can be used to predict the energy consumption of the next sub-path, thereby improving the prediction accuracy of the model for battery energy consumption. In this way, the battery management system can more accurately manage the battery energy of the vehicle to dynamically adjust the energy distribution and management strategy of the vehicle under different driving conditions and extend the cruising range of the vehicle.

[0116] The embodiment of the present application also provides a computer storage medium, which can store multiple instructions, and the instructions are suitable for being loaded and executed by a processor to perform the steps of the method according to any one of the above embodiments.

[0117] Please refer to Figure 9 , Figure 9 which is a schematic structural diagram of a terminal provided by the embodiment of the present application. As Figure 9As shown in the figure, the terminal 900 may include: at least one terminal processor 901, at least one network interface 904, a user interface 903, a memory 905, and at least one communication bus 902.

[0118] Among them, the communication bus 902 is used to realize the connection and communication between these components.

[0119] Among them, the user interface 903 may include a display screen and a camera. Optionally, the user interface 903 may further include a standard wired interface and a wireless interface.

[0120] Among them, the network interface 904 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0121] Among them, the terminal processor 901 may include one or more processing cores. The terminal processor 901 uses various interfaces and lines to connect all parts within the entire terminal 900. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 905, and by calling data stored in the memory 905, it executes various functions of the terminal 900 and processes data. Optionally, the terminal processor 901 may be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), or programmable logic array (PLA). The terminal processor 901 may integrate one or a combination of several of a central processing unit (CPU), a graphics processing unit (GPU), and a modem, etc. Among them, the CPU mainly processes the operating system, user interface, and application programs, etc.; the GPU is responsible for the rendering and drawing of the content to be displayed on the display screen; the modem is used to process wireless communication. It can be understood that the above-mentioned modem may not be integrated into the terminal processor 901 and may be implemented separately through a single chip.

[0122] Among them, the memory 905 may include a Random Access Memory (RAM), or may also include a Read-Only Memory (ROM). Optionally, the memory 905 includes a non-transitory computer-readable storage medium. The memory 905 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 905 may include a program storage area and a data storage area. Among them, the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area may store data involved in the above-mentioned method embodiments. Optionally, the memory 905 may also be at least one storage device located far from the aforementioned terminal processor 901. As Figure 9 shown, the memory 905 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a battery management program.

[0123] In Figure 9 the terminal 900 shown, the user interface 903 is mainly used to provide an input interface for the user to obtain user input data; while the terminal processor 901 can be used to call the battery management program stored in the memory 905 and specifically perform the following operations:

[0124] Obtain the characteristic data of the vehicle on the current sub-path based on the energy consumption correction information of the vehicle on the previous sub-path and the driving information of the vehicle on the current sub-path;

[0125] Input the characteristic data into the battery prediction model to obtain the initial predicted energy consumption of the vehicle on the current sub-path output by the battery prediction model;

[0126] Obtain the actual energy consumption of the vehicle on the current sub-path;

[0127] Based on the characteristic data, the initial predicted energy consumption, and the actual energy consumption, update the prediction error probability distribution of the battery prediction model, and correct the initial predicted energy consumption according to the updated prediction error probability distribution to obtain the energy consumption correction information of the vehicle on the current sub-path.

[0128] In some possible embodiments, the terminal processor 901 further specifically performs the following steps: obtaining the complete path of the vehicle through the map navigation software, and dividing the complete path into multiple sub-paths according to traffic nodes or a preset length; obtaining the characteristic data of the vehicle in the current sub-path based on the energy consumption correction information of the vehicle in the previous sub-path and the driving information of the vehicle in the current sub-path, including: obtaining the traffic characteristics of the current sub-path and the battery characteristics of the vehicle in the current sub-path, and determining the characteristic data of the vehicle in the current sub-path according to the energy consumption correction information, traffic characteristics, and battery characteristics of the vehicle in the previous sub-path.

[0129] In some possible embodiments, when the terminal processor 901 updates the predicted error probability distribution of the battery prediction model based on the characteristic data, the initial predicted energy consumption, and the actual energy consumption, it specifically performs the following steps: calculating the prediction error between the actual energy consumption and the initial predicted energy consumption; based on the prediction error and the characteristic data, updating the prior predicted error probability distribution of the battery prediction model to the posterior predicted error probability distribution according to Bayes' theorem.

[0130] In some possible embodiments, when the terminal processor 901 corrects the initial predicted energy consumption according to the updated predicted error probability distribution to obtain the energy consumption correction information of the vehicle in the current sub-path, it specifically performs the following steps: determining the error correction coefficient according to the expectation of the prediction error in the posterior predicted error probability distribution, and correcting the initial predicted energy consumption based on the error correction coefficient to obtain the corrected predicted energy consumption; using the corrected predicted energy consumption and the error correction coefficient as the energy consumption correction information of the vehicle in the current sub-path.

[0131] In some possible embodiments, the terminal processor 901 further specifically performs the following steps: constructing an initial battery prediction model based on a long short-term memory network and a bidirectional gated recurrent unit; obtaining a plurality of sample characteristic data, and inputting each sample characteristic data into the initial battery prediction model to train the initial battery prediction model; during the training process of the initial battery prediction model, controlling the initial battery prediction model to output the corresponding predicted state of charge based on each sample characteristic data, and adjusting the hyperparameters of the initial battery prediction model according to the predicted state of charge and the true state of charge of each sample characteristic data until the initial battery prediction model converges to obtain the trained battery prediction model.

[0132] In some possible embodiments, when the terminal processor 901 adjusts the hyperparameters of the initial battery prediction model according to the predicted state of charge and the true state of charge of each sample characteristic data until the initial battery prediction model converges, it specifically performs the following steps: calculating the training loss value according to the predicted state of charge and the true state of charge of each sample characteristic data, and adjusting the hyperparameters of the initial battery prediction model based on the training loss value through a metaheuristic algorithm until the initial battery prediction model converges.

[0133] In some possible embodiments, the terminal processor 901 further specifically performs the following steps: obtaining the predicted state of charge output by the battery prediction model, and adjusting the power of the battery on the corresponding sub-path based on the predicted state of charge and the traffic characteristics on the corresponding sub-path.

[0134] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of devices or modules can be in electrical, mechanical or other forms.

[0135] The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical modules, that is, they may be located in one place, or they may be distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0136] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The above computer program product includes one or more computer instructions. When the above computer program instructions are loaded and executed on a computer, the processes or functions described above in accordance with the embodiments of this specification are generated in whole or in part. The above computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The above computer instructions can be stored in a computer-readable storage medium or transmitted through the above computer-readable storage medium. The above computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired manner (such as coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.). The above computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more integrated available media. The above available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a Digital Versatile Disc (DVD)), or a semiconductor medium (for example, a Solid State Disk (SSD)), etc.

[0137] It should be noted that, for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0138] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0139] The above is the description of a battery management method, device, storage medium, and terminal provided by this application. For those skilled in the art, according to the idea of the embodiments of this application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to this application.

Claims

1. A battery management method, characterized in that, The method includes: Obtaining the characteristic data of the vehicle on the current sub-path based on the energy consumption correction information of the vehicle on the previous sub-path and the driving information of the vehicle on the current sub-path; Inputting the characteristic data into the battery prediction model to obtain the initial predicted energy consumption of the vehicle on the current sub-path output by the battery prediction model; Obtaining the actual energy consumption of the vehicle on the current sub-path; Based on the characteristic data, the initial predicted energy consumption, and the actual energy consumption, updating the prediction error probability distribution of the battery prediction model, and correcting the initial predicted energy consumption according to the updated prediction error probability distribution to obtain the energy consumption correction information of the vehicle on the current sub-path.

2. The method according to claim 1, wherein The method further includes: Obtaining the complete path of the vehicle through a map navigation software, and dividing the complete path into multiple sub-paths according to traffic nodes or a preset length; The obtaining the characteristic data of the vehicle on the current sub-path based on the energy consumption correction information of the vehicle on the previous sub-path and the driving information of the vehicle on the current sub-path includes: Obtaining the traffic characteristics of the current sub-path and the battery characteristics of the vehicle on the current sub-path, and determining the characteristic data of the vehicle on the current sub-path according to the energy consumption correction information of the vehicle on the previous sub-path, the traffic characteristics, and the battery characteristics.

3. The method according to claim 1, wherein The updating the prediction error probability distribution of the battery prediction model based on the characteristic data, the initial predicted energy consumption, and the actual energy consumption includes: Calculating the prediction error between the actual energy consumption and the initial predicted energy consumption; Based on the prediction error and the characteristic data, updating the prior prediction error probability distribution of the battery prediction model to a posterior prediction error probability distribution according to Bayes' theorem.

4. The method according to claim 3, wherein The correcting the initial predicted energy consumption according to the updated prediction error probability distribution to obtain the energy consumption correction information of the vehicle on the current sub-path includes: Determining an error correction coefficient according to the expectation of the prediction error in the posterior prediction error probability distribution, and correcting the initial predicted energy consumption based on the error correction coefficient to obtain a corrected predicted energy consumption; Taking the corrected predicted energy consumption and the error correction coefficient as the energy consumption correction information of the vehicle on the current sub-path.

5. The method according to claim 1, wherein The method further includes: Constructing an initial battery prediction model based on a long short-term memory network and a bidirectional gated recurrent unit; Obtaining a plurality of sample characteristic data, and inputting each sample characteristic data into the initial battery prediction model to train the initial battery prediction model; During the training process of the initial battery prediction model, controlling the initial battery prediction model to output corresponding predicted state of charge based on each sample characteristic data, and adjusting the hyperparameters of the initial battery prediction model according to the predicted state of charge and the true state of charge of each sample characteristic data until the initial battery prediction model converges to obtain a trained battery prediction model.

6. The method according to claim 5, wherein The adjusting the hyperparameters of the initial battery prediction model according to the predicted state of charge and the true state of charge of each sample characteristic data until the initial battery prediction model converges includes: Calculate the training loss value according to each predicted state of charge and the true state of charge of each sample feature data, and adjust the hyperparameters of the initial battery prediction model through a metaheuristic algorithm based on the training loss value until the initial battery prediction model converges.

7. The method according to claim 1, wherein The method further includes: Obtain the predicted state of charge output by the battery prediction model, and adjust the power of the battery on the corresponding sub-path based on the predicted state of charge and the traffic characteristics on the corresponding sub-path.

8. A battery management device, characterized in that, The device includes: A feature data acquisition module, configured to acquire the feature data of the vehicle on the current sub-path based on the energy consumption correction information of the vehicle on the previous sub-path and the driving information of the vehicle on the current sub-path; A model prediction module, configured to input the feature data into the battery prediction model to obtain the initial predicted energy consumption of the vehicle on the current sub-path output by the battery prediction model; An actual energy consumption acquisition module, configured to acquire the actual energy consumption of the vehicle on the current sub-path; An error correction module, configured to update the prediction error probability distribution of the battery prediction model based on the feature data, the initial predicted energy consumption, and the actual energy consumption, and correct the initial predicted energy consumption according to the updated prediction error probability distribution to obtain the energy consumption correction information of the vehicle on the current sub-path.

9. A computer storage medium, characterized in that, The computer storage medium stores multiple instructions, and the instructions are adapted to be loaded and executed by a processor to perform the steps of the method according to any one of claims 1 to 7.

10. A terminal, characterized in that, It includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the steps of the method according to any one of claims 1-7.

Images