Vehicle battery optimization suggestion generation method and device based on generative adversarial model, equipment and storage medium

Abstract

The invention discloses a vehicle battery optimization suggestion generation method, device and equipment based on a generative adversarial model, and a storage medium, relates to the technical field of energy management, and discloses a vehicle battery optimization suggestion generation method based on the generative adversarial model, and the method comprises the steps: obtaining the performance parameters of a vehicle uploaded by a vehicle end; preprocessing the performance parameters to obtain preprocessed performance parameters; obtaining a performance parameter weight and a target performance parameter according to the preprocessing performance parameter and a generative adversarial model; and obtaining an optimization suggestion according to the target performance parameter and the performance parameter weight, and sending the optimization suggestion to the vehicle end. Performance parameters of the new energy electric vehicle are analyzed and optimized through the adversarial network model, so that customized optimization suggestions are generated and issued, and the energy use efficiency and the overall performance of the vehicle are improved.

Description

Technical Field

[0001] The present application relates to the field of energy management technology, and in particular to a method, device, equipment and storage medium for generating vehicle battery optimization recommendations based on a generative adversarial model. Background Art

[0002] With the rapid development of new energy electric vehicle technology, the energy efficiency of vehicles has become the key to improving user experience and reducing operating costs. However, traditional methods for optimizing electric vehicle efficiency often rely on simple rules and static data analysis, lack the ability to deeply learn and dynamically adjust real-time data, and fail to fully consider the impact of multi-dimensional factors such as driving behavior, battery status, and ambient temperature, which limits the further improvement of optimization effects.

[0003] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0004] The main purpose of this application is to provide a method, device, equipment and storage medium for generating vehicle battery optimization recommendations based on a generative adversarial model, aiming to solve the technical problem of how to optimize the energy utilization efficiency of new energy electric vehicles in multiple dimensions.

[0005] To achieve the above objectives, the present application proposes a method for generating vehicle battery optimization suggestions based on a generative adversarial model, the method comprising:

[0006] Obtain the vehicle performance parameters uploaded by the vehicle end;

[0007] Preprocessing the performance parameters to obtain preprocessed performance parameters;

[0008] Obtaining performance parameter weights and target performance parameters according to the preprocessing performance parameters and the generative adversarial model;

[0009] An optimization suggestion is obtained according to the target performance parameter and the performance parameter weight, and the optimization suggestion is sent to the vehicle end.

[0010] In one embodiment, the step of obtaining optimization suggestions according to the target performance parameter and the performance parameter weight includes:

[0011] Get the optimization index threshold of the vehicle;

[0012] Obtaining a target optimization index according to the target performance parameter and the performance parameter weight;

[0013] When the target optimization index is greater than the optimization index threshold, obtaining a speed adjustment suggestion, an energy recovery level adjustment suggestion, and an acceleration adjustment suggestion according to the target optimization index;

[0014] Obtaining a comprehensive optimization suggestion according to the speed adjustment suggestion, the energy recovery level adjustment suggestion, and the acceleration adjustment suggestion;

[0015] The comprehensive optimization suggestion is formatted to obtain an optimization suggestion.

[0016] In one embodiment, when the target optimization index is greater than the optimization index threshold, the step of obtaining a speed adjustment suggestion, an energy recovery level adjustment suggestion, and an acceleration adjustment suggestion according to the target optimization index includes:

[0017] When the target optimization index is greater than the optimization index threshold, obtaining the current speed, speed threshold, current energy recovery efficiency, energy recovery efficiency threshold, current energy recovery level, preset energy level, current acceleration and acceleration threshold of the vehicle;

[0018] When the current speed is greater than the speed threshold, a target speed is obtained according to the current speed and the target optimization index, and a speed adjustment suggestion is obtained according to the target speed;

[0019] When the current energy recovery efficiency is less than the energy recovery efficiency threshold and the current energy recovery level is less than the preset energy level, a target recovery level is obtained according to the current energy recovery level and the target optimization index, and an energy recovery level adjustment suggestion is obtained according to the target recovery level;

[0020] When the current acceleration is greater than the acceleration threshold, a target acceleration is obtained according to the current acceleration and the target optimization index, and an acceleration adjustment suggestion is obtained according to the target acceleration.

[0021] In one embodiment, the step of preprocessing the performance parameter to obtain the preprocessed performance parameter includes:

[0022] Identify missing fields of the performance parameters, and fill the missing fields according to an average value of historical data of the vehicle or an average value of the missing fields to obtain complete performance parameters;

[0023] Obtaining a first mean value and a first standard deviation of the complete performance parameter and a second mean value and a second standard deviation of the vehicle set parameter;

[0024] Obtaining a standard performance parameter according to at least one of the complete performance parameter, the first average value, the first standard deviation, the second average value, and the second standard deviation;

[0025] The standard performance parameters are sorted according to the upload time, and the sorted standard performance parameters are divided according to the preset time window to obtain the preprocessing performance parameters.

[0026] In one embodiment, the step of obtaining a standard performance parameter according to at least one of the complete performance parameter, the first average value, the first standard deviation, the second average value, and the second standard deviation comprises:

[0027] Obtaining a normal value range of a complete performance parameter according to the first mean value and the first standard deviation;

[0028] Obtaining normal performance parameters according to the complete performance parameters and the normal value range;

[0029] Obtaining a local standardized parameter according to the normal performance parameter, the first average value, and the first standard deviation;

[0030] A global standardized parameter is obtained according to the local standardized parameter, the second average value and the second standard deviation, and a standard performance parameter is obtained according to the global standardized parameter.

[0031] In one embodiment, before the step of obtaining the performance parameter weights and the target performance parameters according to the preprocessing performance parameters and the generative adversarial model, the step further includes:

[0032] Obtain initial performance parameter weights, the generator of the adversarial model, and the discriminator of the adversarial model;

[0033] According to the generator and the discriminator, a generator loss function and a discriminator loss function are obtained;

[0034] Optimizing the initial performance parameter weights according to the generator loss function to obtain target performance parameter weights;

[0035] Update the generative adversarial model according to the generator loss function, the discriminator loss function and the preset optimizer;

[0036] When the generator output of the updated generative adversarial model is the target performance indicator, the updating is terminated to obtain the updated generative adversarial model.

[0037] In one embodiment, the step of optimizing the initial performance parameter weight according to the generator loss function to obtain the target performance parameter weight includes:

[0038] Obtaining a weight parameter gradient according to the generator loss function and the initial performance parameter weight;

[0039] Obtaining an optimized performance parameter weight according to the initial performance parameter weight, the weight parameter gradient and the preset learning rate;

[0040] When the weight parameter gradient is equal to a preset gradient value or the generator loss function is a preset output, the target performance parameter weight is obtained according to the optimized performance parameter weight.

[0041] In addition, to achieve the above purpose, the present application also proposes a vehicle battery optimization suggestion generation device based on a generative adversarial model, and the vehicle battery optimization suggestion generation device based on a generative adversarial model includes: an acquisition module, used to acquire the performance parameters of the vehicle uploaded by the vehicle end;

[0042] A processing module, used for preprocessing the performance parameters to obtain preprocessed performance parameters;

[0043] An obtaining module, used for obtaining performance parameter weights and target performance parameters according to the preprocessing performance parameters and the generative adversarial model;

[0044] The optimization module is used to obtain optimization suggestions based on the target performance parameters and the performance parameter weights, and send the optimization suggestions to the vehicle end.

[0045] In addition, to achieve the above-mentioned purpose, the present application also proposes a vehicle battery optimization recommendation generation device based on a generative adversarial model, the device comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the vehicle battery optimization recommendation generation method based on a generative adversarial model as described above.

[0046] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the method for generating vehicle battery optimization recommendations based on the generative adversarial model as described above are implemented.

[0047] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the vehicle battery optimization recommendation generation method based on the generative adversarial model as described above.

[0048] One or more technical solutions proposed in this application have at least the following technical effects:

[0049] The technical means of obtaining the performance parameters of the vehicle uploaded by the vehicle side; preprocessing the performance parameters to obtain preprocessed performance parameters; obtaining performance parameter weights and target performance parameters based on the preprocessed performance parameters and a generative adversarial model; obtaining optimization suggestions based on the target performance parameters and the performance parameter weights, and sending the optimization suggestions to the vehicle side solves the problem of optimizing the energy utilization efficiency of new energy electric vehicles. Compared with the existing technology, more accurate performance prediction and more effective energy management are achieved, thereby improving the energy utilization efficiency and overall performance of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0051] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0052] Figure 1 A flowchart diagram of a first embodiment of a method for generating vehicle battery optimization suggestions based on a generative adversarial model in this application;

[0053] Figure 2 A flow chart of a second embodiment of a method for generating vehicle battery optimization suggestions based on a generative adversarial model in this application;

[0054] Figure 3 A schematic diagram of the module structure of a vehicle battery optimization suggestion generating device based on a generative adversarial model according to an embodiment of the present application;

[0055] Figure 4 Schematic diagram of the device structure of the hardware operating environment involved in the method for generating vehicle battery optimization recommendations based on a generative adversarial model in an embodiment of the present application.

[0056] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0057] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0058] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0059] The main solution of the embodiment of the present application is: obtaining the performance parameters of the vehicle uploaded by the vehicle end; preprocessing the performance parameters to obtain preprocessed performance parameters; obtaining performance parameter weights and target performance parameters based on the preprocessed performance parameters and a generative adversarial model; obtaining optimization suggestions based on the target performance parameters and the performance parameter weights, and sending the optimization suggestions to the vehicle end.

[0060] In this embodiment, for ease of description, the following description is made by taking the identification of the cloud computing platform as the execution subject.

[0061] Since the existing technology cannot optimize the energy utilization efficiency of new energy electric vehicles in multiple dimensions, the present application provides a solution, which adopts the method of obtaining the performance parameters of the vehicle uploaded by the vehicle side; preprocessing the performance parameters to obtain preprocessed performance parameters; obtaining performance parameter weights and target performance parameters based on the preprocessed performance parameters and a generative adversarial model; obtaining optimization suggestions based on the target performance parameters and the performance parameter weights, and sending the optimization suggestions to the vehicle side. This technical means solves the problem of optimizing the energy utilization efficiency of new energy electric vehicles. Compared with the existing technology, it achieves more accurate performance prediction and more effective energy management, thereby improving the energy utilization efficiency and overall performance of the vehicle.

[0062] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, a vehicle battery optimization suggestion generation device based on a generative adversarial model, and a cloud computing platform, etc. The following takes the cloud computing platform as an example to illustrate this embodiment and the following embodiments.

[0063] Based on this, the embodiment of the present application provides a method for generating vehicle battery optimization suggestions based on a generative adversarial model, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the method for generating vehicle battery optimization recommendations based on a generative adversarial model in the present application.

[0064] In this embodiment, the vehicle battery optimization suggestion generation method based on the generative adversarial model includes steps S10 to S40:

[0065] Step S10, obtaining the performance parameters of the vehicle uploaded by the vehicle end;

[0066] It should be noted that the vehicle side refers to the hardware and software systems on the vehicle, including various sensors, controllers, computing units, etc. These devices are responsible for real-time monitoring of the vehicle's operating status and collecting relevant data.

[0067] In addition, it should be noted that the vehicle's performance parameters are quantitative indicators of the battery status when the vehicle is in operation, including but not limited to obtaining data such as the battery's charging speed (C) and health status (SOH) through the battery management system; obtaining data such as the ambient temperature (T_c) through environmental sensors; and obtaining the charging frequency (Ch) through vehicle usage pattern analysis.

[0068] Step S20, preprocessing the performance parameters to obtain preprocessed performance parameters;

[0069] It should be noted that preprocessing is a series of processing steps performed on raw data to improve the quality and usability of the data. Preprocessing performance parameters refer to the data after the preprocessing steps, which are ready for modeling and analysis and can be used for subsequent steps.

[0070] In a feasible implementation, step S20 may include steps S21 to S25:

[0071] Step S21, identifying the missing fields of the performance parameters, and filling the missing fields according to the average value of the historical data of the vehicle or the average value of the missing fields to obtain complete performance parameters;

[0072] It should be noted that a missing field refers to an item that lacks data in a performance parameter data set. For example, if the running speed data at a certain point in time is not successfully recorded or transmitted, then the "running speed" field is considered missing in the performance parameter data transmitted at that point in time.

[0073] In addition, it should be noted that the average value of historical data refers to the average value of historical uploaded data of a certain performance parameter of the current vehicle within a period of time. The history here is relative to the current time point, and the data uploaded before the current time point is considered as historical data. The average value of missing fields is the average of all known values ​​for a specific field.

[0074] It can be understood that when the average value of the vehicle's historical data exists and can be queried, the missing field is filled with the average value of the vehicle's historical data; when the average value of the vehicle's historical data does not exist or cannot be queried, the missing field is filled with the average value of the missing field. Missing data may lead to inaccurate analysis results or model training failure. By identifying and filling missing values ​​in the data set, the integrity and availability of the data are ensured. By filling with the average value of historical data or the average value of the field, the impact of missing data on the overall analysis can be effectively reduced.

[0075] Step S22, obtaining a first mean value and a first standard deviation of the complete performance parameter and a second mean value and a second standard deviation of the vehicle set parameter;

[0076] It should be noted that the first average value refers to the mean value obtained by statistically analyzing the performance parameters of a single vehicle. For example, if the performance parameters include battery charging speed, the first average value does not represent the average value of all charging speed records of this vehicle within a week. The first standard deviation represents the degree of variation of the performance parameters of a single vehicle. It measures the difference between the performance parameter value of a single vehicle and its first average value. A larger standard deviation means that the data points are more widely distributed and the difference is greater.

[0077] In addition, it should be noted that the vehicle set parameter represents a set of performance parameters of multiple vehicles. The second average value is the average value of a specific performance parameter of all vehicles. For example, if the vehicle set parameter includes the battery charging speed of all vehicles, then the second average value is the overall average value of all charging speed records. The second standard deviation represents the degree of variation of the performance parameters of all vehicles. It measures the difference between the vehicle set parameter value and the second average value.

[0078] Step S23, obtaining a standard performance parameter according to at least one of the complete performance parameter, the first average value, the first standard deviation, the second average value, and the second standard deviation;

[0079] It should be noted that the standard performance parameters are performance parameters that have been processed for outliers and normalized. The purpose of the outlier processing is to remove data outside the standard range, and the normalization processing generally includes local normalization and / or global normalization.

[0080] Local standardization and global standardization are data preprocessing methods that are performed on a single vehicle data set and a multi-vehicle data set, respectively. Local standardization helps eliminate the dimensionality effect within a single data set, while global standardization further ensures the comparability between different vehicle data sets. The output of local standardization and / or global standardization can be directly used for subsequent data analysis and machine learning model training, thereby improving the performance and accuracy of the model.

[0081] In a feasible implementation, step S23 may include steps S231 to S234:

[0082] Step S231, obtaining a normal value range of a complete performance parameter according to the first mean value and the first standard deviation;

[0083] It should be noted that the normal value range refers to the expected fluctuation range of the performance parameter determined based on statistical analysis, and this range is usually used to identify and process outliers, that is, data points that significantly deviate from the normal range. In this embodiment, the minimum value of the normal value range is the first mean value minus 3 times the first standard deviation, and the maximum value of the normal value range is the first mean value plus 3 times the first standard deviation. The first mean value ± 3 times the first standard deviation can cover about 99.7% of the data points, which is usually considered to be the critical point of outliers.

[0084] In the specific implementation process, suppose there is a data set that records the battery charging speed (unit: kWh) of 7 vehicles in the same time window, and the data is as follows: {Day1:50; Day2:45; Day3:48; Day4:47; Day5:50; Day6:46; Day7:109}; the calculated mean is 53.57 (rounded to two decimal places); the calculated standard deviation is 18.3; the normal maximum value is 53.57+18.3*3=108.47; the normal minimum value is 53.57-18.3*3=-1.33, which is approximately equal to 0. The normal value range is 0 to 108.47.

[0085] Step S232, obtaining normal performance parameters according to the complete performance parameters and the normal value range;

[0086] It is understood that for each data point in the complete performance parameter, it is checked whether it falls within the normal value range. If the data points are beyond the normal value range, these data points are marked as outliers and may be removed in subsequent steps or replaced according to the mean replacement step of step S21. After removing the outliers, all data points that fall within the normal value range are confirmed, and these data points are considered to be normal performance parameters.

[0087] It should be noted that normal performance parameters refer to data points that fall within the normal value range. These parameters are considered to be the performance of the vehicle under normal operating conditions and are suitable for further data analysis and model training.

[0088] Step S233, obtaining a local standardized parameter according to the normal performance parameter, the first average value and the first standard deviation;

[0089] It is understandable that the data of the corresponding field of a certain vehicle needs to be locally standardized. Locally standardized data is usually used in scenarios where individual differences need to be considered. For example, when the performance or usage conditions of different vehicles vary significantly, local standardization can help the model better understand and predict the specific behavior or state of each vehicle. The formula for local standardization is as follows:

[0090]

[0091] Where x is the value of the normal performance parameter, μ local is the first mean value, σ local is the first standard deviation, z local The data are locally normalized.

[0092] Step S234, obtaining a global standardization parameter according to the local standardization parameter, the second mean value and the second standard deviation, and obtaining a standard performance parameter according to the global standardization parameter.

[0093] It is understandable that global normalization is required for the overall data of all vehicles. Globally normalized data is usually used in scenarios where differences between different vehicles need to be eliminated for cross-vehicle comparison or analysis. For example, when the average performance or trend of the entire fleet needs to be evaluated, global normalization can make the data comparable. The formula for global normalization is as follows:

[0094]

[0095] Where x is the local normalized value, μ global is the second mean value, σ global is the second standard deviation, z global The data is globally standardized.

[0096] Step S24, sorting the standard performance parameters according to the upload time, and dividing the sorted standard performance parameters according to the preset time window to obtain pre-processing performance parameters.

[0097] It is understandable that since the vehicle performance parameters may be uploaded at different time points, they need to be sorted according to the actual upload time of these parameters. This is done to maintain the temporal integrity of the data and ensure the accuracy of the analysis.

[0098] It should be noted that the preset time window refers to a fixed time period used to divide the data set in data analysis. In this embodiment, the preset time window is 5 minutes, and the data will be divided into blocks of 5 minutes each. This division helps to analyze the changes in vehicle performance within a specific time period and can also improve the efficiency of data analysis.

[0099] It can be understood that the preprocessing performance parameters are the performance parameters after data missing value identification and processing, outlier processing, standardization processing and time division, and these performance parameters can be used for subsequent analysis of the model.

[0100] Step S30, obtaining performance parameter weights and target performance parameters according to the preprocessing performance parameters and the generative adversarial model;

[0101] It should be noted that the generative adversarial model refers to the combination of the generative adversarial network (GAN), the loss function and the optimizer. The generative adversarial network is a special type of neural network consisting of two parts, including a generator responsible for generating data and a discriminator responsible for evaluating the authenticity of the data. The loss function, also known as the cost function, is used to evaluate the difference or loss between the model's predicted value and the true value. The loss function is actually a mathematical function that takes the model's predicted value and the true value as input and outputs a non-negative real number to represent the size of the loss. The optimizer is used to update the model's parameters during the model training process to minimize the loss function. The optimizer calculates the gradient of the loss function relative to the model parameters and adjusts the parameters based on these gradients. The gradient adjustment affects the efficiency and final performance of the model training.

[0102] In this embodiment, GAN is used to simulate and optimize the performance parameters of the vehicle, and binary cross entropy (BCE) is selected as the loss function and Adam optimizer is used as the optimizer. BCE helps the discriminator of GAN to distinguish between real data and fake data generated by the generator of GAN; at the same time, Adam optimizer is used to automatically adjust the weight parameters of the generator and discriminator according to the gradient of the loss function to minimize the loss and improve the model performance, thereby realizing the effective simulation and optimization of vehicle performance parameters.

[0103] Binary cross entropy (BCE) as a loss function can directly measure the probability difference. Binary cross entropy directly measures the difference between the probability distribution of the model output and the true label. By minimizing the cross entropy loss, the output probability distribution of the model can be made closer to the true label, thereby improving the prediction accuracy of the model. Gradient descent optimization can be performed. The binary cross entropy loss function is differentiable, which means that optimization algorithms such as gradient descent can be used to minimize the loss function. This allows the model to gradually adjust parameters during training to better fit the data. It can handle unbalanced data. In binary classification problems, if the number of positive and negative samples is extremely unbalanced, the binary cross entropy loss function can give a larger weight to minority class samples, thereby alleviating the impact of unbalanced data on model performance to a certain extent. It can be updated online and converge quickly. The binary cross entropy loss function allows the model to be updated online, that is, it can process new data without retraining the entire model. In addition, due to its optimization characteristics, the model can usually converge to the optimal solution faster. Binary cross entropy (BCE) as a loss function is often used in classification problems because of its intuitiveness and computational simplicity.

[0104] Adam optimizer is used for model training: learning rate can be adjusted adaptively: Adam combines the advantages of AdaGrad and RMSProp optimization algorithms and can adaptively adjust the learning rate, which is very useful when dealing with non-stationary objective functions. Adam usually requires less memory and can optimize efficiently. At the same time, the Adam optimizer is suitable for big data and high-dimensional space. When faced with high-dimensional parameter space or large-scale data sets, Adam often performs better. Compared with other optimization algorithms, Adam is less likely to fall into local optimal solutions, especially in complex, high-dimensional optimization problems.

[0105] In addition, it should be noted that the performance parameter weight refers to the relative importance of each performance parameter (such as battery charging speed, battery health status, etc.) during the optimization process. These weights are learned by the GAN model and are used to determine which parameters should be prioritized in performance optimization. The target performance parameters are the performance goals set for the vehicle after GAN model analysis and optimization. These goals reflect the ideal state of vehicle performance under current operating conditions and are used to guide vehicle operation and maintenance.

[0106] Step S40, obtaining optimization suggestions based on the target performance parameters and the performance parameter weights, and sending the optimization suggestions to the vehicle end.

[0107] It should be noted that optimization suggestions are a series of operational suggestions based on target performance parameters and performance parameter weights, which are intended to guide the vehicle to adjust its operating parameters to approach or reach the target performance state. These suggestions may include adjusting the battery charging strategy, changing the driving mode, optimizing the energy recovery system, etc.

[0108] It is understandable that sending to the vehicle side involves transmitting the optimization suggestions to the vehicle's control system via a wireless network. The sending operation is usually implemented through the vehicle's communication interface, such as an on-board information system or a vehicle networking module, and this embodiment does not limit this.

[0109] In the specific implementation process, a certain model of new energy electric vehicle is running on a city road at a speed of 60km / h and an acceleration of 0.2m / s 2, and its energy efficiency reaches 12kWh per 100 kilometers. The on-board sensors collect the vehicle's driving data in real time, including speed, acceleration and energy consumption, and upload them to the cloud platform. The model of the cloud platform receives the data uploaded by the on-board sensors and performs real-time analysis. By comparing the real-time data with the pre-set efficiency threshold, the model finds that the current energy efficiency is lower than the expected level. Based on the analysis results of the machine learning model, the cloud platform generates targeted optimization suggestions to improve the energy efficiency of new energy electric vehicles. Taking into account the current vehicle speed and acceleration, it is recommended that the driver reduce the speed to 55km / h and adjust to a more economical driving mode or energy recovery level. The optimization suggestions are sent to the vehicle information system via the wireless network, and the vehicle information system displays the suggestions to the driver in a friendly way, such as through the central control screen prompts or voice prompts, to guide the driver to adopt more energy-saving driving behaviors.

[0110] In a feasible implementation, step S40 may include steps S41 to S45:

[0111] Step S41, obtaining the optimization index threshold of the vehicle;

[0112] It should be noted that the optimization index is based on the concept of multi-objective optimization. In the context of optimizing the efficiency of new energy electric vehicles, we not only focus on the state of charge (SOC) and health state (SOH) of the battery, but also factors such as charging speed (C), ambient temperature (T_c), and charging frequency (Ch). These indicators may point to different optimization directions and have different dimensions. Therefore, it is necessary to comprehensively consider these key parameters to define a comprehensive optimization goal, which is the optimization index. The optimization index threshold refers to a series of standards or critical values ​​used to evaluate whether the vehicle performance has reached the expected optimization target. This threshold is usually set based on the vehicle's design parameters, performance requirements, safety standards, and operating conditions, and can also be calibrated by calculating the historical target optimization index.

[0113] Step S42, obtaining a target optimization index according to the target performance parameter and the performance parameter weight;

[0114] It can be understood that the target optimization index is derived based on the target performance parameter and the performance parameter weight. The larger the target optimization index, the better the overall performance and health status of the battery. The target optimization index is used to guide the specific goals of vehicle performance optimization. The calculation formula of the target optimization index is as follows:

[0115] O′=ω·C+η·H+theta·T c +iota·S+kappa·Ch (Formula 3)

[0116] In the formula, C is the charging speed of the battery, reflecting the charging efficiency and time cost; H is the battery health status, reflecting the long-term reliability of the battery; T c is the ambient temperature during charging, which affects the charging and discharging efficiency and life of the battery; S is the state of charge of the battery, Ch is directly related to the driving range and performance; is the charging frequency, which affects the cycle life and overall maintenance cost of the battery, and O′ is the target optimization index. ω, η, iota and kappa are weight parameters for the optimization scenario of the use efficiency of new energy electric vehicles, which determine the relative importance of each index in the optimization target.

[0117] Step S43, when the target optimization index is greater than the optimization index threshold, obtaining a speed adjustment suggestion, an energy recovery level adjustment suggestion, and an acceleration adjustment suggestion according to the target optimization index;

[0118] It can be understood that the target optimization index is greater than the optimization index threshold, and the energy utilization efficiency of the target optimization index calculated is greater than the optimization index threshold that can be optimized, indicating that adjusting the vehicle according to the calculated target optimization index can beneficially optimize the vehicle.

[0119] It should be noted that speed adjustment suggestions refer to suggestions on adjusting the vehicle's driving speed based on the target optimization indicators and current vehicle performance. Speed ​​adjustment suggestions may involve increasing or decreasing speed to improve energy efficiency. Energy recovery level adjustment suggestions refer to suggestions on adjusting the working level of the vehicle's energy recovery system based on the target optimization indicators. The energy recovery system has different working levels, and adjusting these levels can optimize energy recovery efficiency. Acceleration adjustment suggestions refer to suggestions on adjusting the vehicle's acceleration based on the target optimization indicators. Adjusting the acceleration helps improve the vehicle's energy efficiency and driving safety.

[0120] In a feasible implementation, step S43 may include steps S431 to S434:

[0121] Step S431, when the target optimization index is greater than the optimization index threshold, obtaining the current speed, speed threshold, current energy recovery efficiency, energy recovery efficiency threshold, current energy recovery level, preset energy level, current acceleration and acceleration threshold of the vehicle;

[0122] It should be noted that the current speed of the vehicle refers to the actual driving speed of the vehicle at the current point in time. The speed threshold is a preset speed standard used to evaluate whether the current speed of the vehicle is within the desired range. The current energy recovery efficiency refers to the actual efficiency of the vehicle's energy recovery system at the current point in time. The energy recovery efficiency threshold is a preset energy recovery efficiency standard used to evaluate whether the vehicle's energy recovery efficiency reaches the desired level. The current energy recovery level refers to the current working level of the vehicle's energy recovery system, which is usually divided into multiple levels, such as weak, standard, and strong. The preset energy level refers to the energy recovery level preset according to the vehicle's performance requirements and operating conditions. The current acceleration refers to the actual acceleration of the vehicle at the current point in time. The acceleration threshold is a preset acceleration standard used to evaluate whether the vehicle's current acceleration is within the desired range.

[0123] It is understandable that, with the target optimization index as the optimization target, multiple sub-indicators can be obtained, for example, sub-indicator 1 focuses on charging efficiency, sub-indicator 2 focuses on health status, etc. The weights of the sub-indicators are determined by the user's personal usage habits, vehicle type, and environmental factors. The target optimization index in steps S432 to S434 can be a sub-indicator.

[0124] Step S432, when the current speed is greater than the speed threshold, obtaining a target speed according to the current speed and the target optimization index, and obtaining a speed adjustment suggestion according to the target speed;

[0125] It should be noted that the target speed is a recommended speed calculated based on the current speed and the target optimization index. This speed value is part of the optimization suggestion and is intended to guide the vehicle to adjust its driving speed to meet a specific performance goal.

[0126] It can be understood that the target speed refers to the speed value that minimizes the target optimization index. The calculation formula of the target speed is as follows:

[0127] V optimal =argmin V O1 (Formula 4)

[0128] Where V optimal is the target speed, and O represents the target optimization index.

[0129] Step S433, when the current energy recovery efficiency is less than the energy recovery efficiency threshold and the current energy recovery level is less than the preset energy level, a target recovery level is obtained according to the current energy recovery level and the target optimization index, and an energy recovery level adjustment suggestion is obtained according to the target recovery level;

[0130] It should be noted that the target recovery level is a recommended energy recovery level calculated based on the current energy recovery level and the target optimization index. This level is part of the optimization recommendation and is intended to guide the vehicle to adjust the working state of its energy recovery system to meet specific performance goals.

[0131] In this embodiment, the preset capacity energy level is strong. The current energy recovery level is less than the preset energy level, indicating that the energy recovery level is weak or standard.

[0132] It can be understood that the target recovery level refers to the energy recovery level that minimizes the target optimization index. The calculation formula of the target recovery level is as follows:

[0133] G optimal =argmin G O2 (Formula 5)

[0134] In the formula, G optimal is the target recovery level, and O2 is the target optimization index.

[0135] Step S434, when the current acceleration is greater than the acceleration threshold, a target acceleration is obtained according to the current acceleration and the target optimization index, and an acceleration adjustment suggestion is obtained according to the target acceleration.

[0136] It should be noted that the target acceleration is a recommended acceleration calculated based on the current acceleration and the target optimization index. This acceleration value is part of the optimization suggestion and is intended to guide the vehicle to adjust its acceleration to meet a specific performance goal.

[0137] In this embodiment, the current acceleration is greater than the acceleration threshold, which means that the absolute value of the current acceleration is greater than the acceleration threshold.

[0138] It can be understood that the target acceleration refers to the acceleration magnitude that minimizes the target optimization index. The target acceleration is calculated as follows:

[0139] A optimal =argmin A O3 (Formula 6)

[0140] In the formula, A optimal is the target acceleration, and O3 is the target optimization index.

[0141] Step S44, obtaining a comprehensive optimization suggestion according to the speed adjustment suggestion, the energy recovery level adjustment suggestion and the acceleration adjustment suggestion;

[0142] It should be noted that comprehensive optimization suggestions refer to integrating the adjustment suggestions from all aspects mentioned above to form a comprehensive optimization plan to ensure that the vehicle can maximize energy efficiency while meeting performance requirements.

[0143] It can be understood that the comprehensive optimization suggestion is represented by a vector, and the formula of the comprehensive optimization suggestion is as follows:

[0144]

[0145] In the formula, For comprehensive optimization suggestions, S1, S2 and S3 are speed adjustment suggestions, energy recovery level adjustment suggestions and acceleration adjustment suggestions respectively. ω1, ω2 and ω3 are the weights of speed adjustment suggestions, energy recovery level adjustment suggestions and acceleration adjustment suggestions respectively.

[0146] In addition, it is understandable that the weight in Formula 7 can be assigned to each sub-indicator through expert knowledge, historical data analysis, or machine learning model prediction to reflect its relative importance to the target optimization indicator.

[0147] Step S45, converting the format of the comprehensive optimization suggestion to obtain the optimization suggestion.

[0148] It is understandable that format conversion refers to converting the comprehensive optimization suggestions into a specific data format or expression method to facilitate display, transmission and execution on different systems or devices. In this embodiment, the comprehensive optimization suggestions in vector format need to be converted into JSON objects.

[0149] It should be noted that the optimization suggestion is the data after the comprehensive optimization suggestion is converted into a format. In this embodiment, the comprehensive optimization suggestion in vector format is converted into a JSON object, and the JSON object is sent to the vehicle information system through a secure data transmission protocol, such as HTTPS. The vehicle information system parses the received JSON object and displays it to the driver in various ways such as charts, text or voice. If the vehicle supports an intelligent management system, the optimization suggestion can also be directly executed.

[0150] This embodiment provides a method for generating vehicle battery optimization suggestions based on a generative adversarial model, which solves the technical problem of optimizing the energy efficiency of new energy electric vehicles during operation by using a generative adversarial network (GAN) model and multi-objective optimization technology. By collecting vehicle performance parameters in real time, performing data preprocessing and outlier processing, and simulating and optimizing vehicle performance parameters using a GAN model, the energy efficiency of new energy electric vehicles can be accurately predicted and improved. In addition, by setting optimization index thresholds and weight parameters, targeted optimization suggestions are generated, such as speed adjustment, energy recovery system adjustment, and acceleration adjustment, to guide the vehicle to adjust operating parameters to approach or reach the target performance state. These optimization suggestions are sent to the vehicle end in a friendly manner to guide the driver to adopt a more energy-saving driving behavior, or are directly executed by the vehicle's intelligent management system, thereby achieving the beneficial effect of improving the energy efficiency and overall performance of new energy electric vehicles.

[0151] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can be referred to the above introduction, and will not be repeated in the following. Figure 2 Before step S30, the vehicle battery optimization suggestion generation method based on the generative adversarial model further includes steps S25 to S29:

[0152] Step S25, obtaining initial performance parameter weights, a generator of the adversarial model, and a discriminator of the adversarial model;

[0153] It should be noted that the initial performance parameter weights are randomly assigned values ​​to the generator and discriminator before starting training. These weights will be continuously updated based on the training data to improve the performance of the model.

[0154] In addition, it should be noted that the generator is used to generate fake data samples, with input nodes equal to the number of input features. The generator input is a random noise vector and conditional information (such as battery health status, ambient temperature). The generator output is the predicted corresponding battery performance indicator. The discriminator is used to distinguish between real data and generated fake data, with input nodes equal to the number of input features. The discriminator input is the real corresponding battery performance indicator and the fake corresponding battery performance indicator generated by the generator. The discriminator output is a probability estimate of the input data being the real corresponding battery performance indicator.

[0155] It is understandable that the generator and discriminator of the generative adversarial network constantly compete with each other, so that ultimately, by inputting the current battery data, sufficiently realistic performance indicators can be generated.

[0156] In addition, it can be understood that the number of input nodes equal to the number of input features means that in the architecture of the neural network, each input feature corresponds to an input node. This means that if there is a sample with 5 features, then the input layer of the network will have 5 input nodes, each node receiving a specific feature value.

[0157] Step S26, obtaining a generator loss function and a discriminator loss function according to the generator and the discriminator;

[0158] It is understandable that the generator loss function calculates the error rate of the discriminator's judgment on the data generated by the generator, that is, it is hoped that the discriminator will mistake the generated data for real data. The generator loss function is as follows:

[0159]

[0160] In the formula, G(z i ) is the noise of the generator input, z i is the prediction data sample generated by the generator, N is the number of samples, D(G(z i )) is the discriminator’s decision on the generator input noise.

[0161] In addition, it can be understood that the discriminator loss function is as follows:

[0162]

[0163] In the formula, y i is the true label of the i-th sample, which takes the value of 1 or 0, where 1 represents true and 0 represents false; D(x i ) is the discriminator for the real sample x i The predicted probability of ; N is the number of samples.

[0164] Step S27, optimizing the initial performance parameter weights according to the generator loss function to obtain target performance parameter weights;

[0165] It should be noted that the target performance parameter weights refer to the parameter values ​​of the generator and discriminator after optimization through the training process. These weights enable the model to better complete its task.

[0166] It can be understood that the optimization process refers to the process of adjusting the model weights through an iterative algorithm, such as the gradient descent algorithm.

[0167] Step S28, updating the generative adversarial model according to the generator loss function, the discriminator loss function and the preset optimizer;

[0168] It is understandable that the optimizer is an algorithm for updating model parameters, which adjusts the parameters according to the gradient of the loss function. In this embodiment, the preset optimizer is the Adam optimizer.

[0169] The parameters of the generator or discriminator can be optimized using the Adam optimizer. The optimization process is as follows:

[0170] Calculate the gradient of the loss function. The gradient calculation formula of the loss function is as follows:

[0171]

[0172] In the formula, θ is the parameter of the generator or discriminator, L can be the loss function of the generator or the loss function of the discriminator, and g t is the gradient of the loss function.

[0173] Calculate the exponentially weighted moving average of the gradient, that is, the first-order moment estimate. The formula for the first-order moment estimate is as follows:

[0174] m t =β1×m t-1 +(1-β1)×g t (Formula 11)

[0175] In the formula, β1 is a hyperparameter, g t is the gradient of the loss function, m t is the first-order moment estimate at time t, m t-1 is the first-order moment estimate at time t-1.

[0176] Calculate the exponentially weighted moving average of the squared gradient, that is, the second-order moment estimate. The formula for the second-order moment estimate is as follows:

[0177] v t =β2×v t-1 +(1-β2)×g t 2 (Formula 12)

[0178] In the formula, β2 is a hyperparameter, g t is the gradient of the loss function, v t is the second-order moment estimate at time t, v t-1 is the second-order moment estimate at time t-1.

[0179] Calculate the bias correction of the first-order moment estimate. The calculation formula for the bias correction of the first-order moment estimate is as follows:

[0180]

[0181] In the formula, is the first-order moment estimation bias correction, m tis the first-order moment estimate at time t, and β1 is a hyperparameter.

[0182] Calculate the bias correction of the second-order moment estimate. The calculation formula for the bias correction of the second-order moment estimate is as follows:

[0183]

[0184] In the formula, is the bias correction of the second-order moment estimate, v t is the second-order moment estimate at time t, and β2 is a hyperparameter.

[0185] Update the parameters of the generator or discriminator. The formula for updating the parameters is as follows:

[0186]

[0187] In the formula, θ is the parameter of the generator or discriminator, α is the learning rate, is the first-order moment estimation bias correction, is the bias correction for the second-order moment estimate, ∈ is a very small number used to prevent the denominator from being zero.

[0188] It can be understood that through the final output θ t , which is the parameter of the generator or discriminator, can update the generative adversarial model.

[0189] In a feasible implementation, step S28 may include steps S281 to S283:

[0190] Step S281, obtaining a weight parameter gradient according to the generator loss function and the initial performance parameter weight;

[0191] It should be noted that the weight parameter gradient refers to the derivative of the loss function with respect to the model weights. It is a vector that indicates the direction in which the loss function increases fastest in the weight space. In the gradient descent algorithm, the gradient is used to guide the adjustment of the weights to minimize the loss function. The direction of the weight update is usually the opposite direction of the gradient because this is the direction of reducing the loss.

[0192] Taking the performance parameter weight ω as an example, the process of updating the performance weight parameter is explained. η, iota and kappa can all be substituted into the following steps.

[0193] It can be understood that the calculation formula of the weight parameter gradient is as follows:

[0194]

[0195] Where L is the output of the generator loss function, ω is the performance parameter weight of the battery charging speed, and g ω is the weight parameter gradient.

[0196] Step S282, obtaining an optimized performance parameter weight according to the initial performance parameter weight, the weight parameter gradient and the preset learning rate;

[0197] It should be noted that the optimization performance parameter weights are weights updated according to the gradient descent algorithm, aiming to reduce the value of the loss function.

[0198] It can be understood that the calculation formula for optimizing the performance parameter weight is as follows:

[0199] ω new =ω old -α×g ω (Formula 17)

[0200] In the formula, ω new is the optimization performance parameter weight, is the initial performance parameter weight, α is the learning rate, g ω is the weight parameter gradient.

[0201] Step S283, when the weight parameter gradient is equal to a preset gradient value or the generator loss function is a preset output, the target performance parameter weight is obtained according to the optimized performance parameter weight.

[0202] It should be noted that the preset output is a target value of the generator loss function, which is used to determine whether the model has achieved the expected performance. When the generator loss function reaches the preset output, it means that the generator has generated sufficiently realistic data, and training can be stopped at this time. The target performance parameter weight is the final weight value that is expected to be achieved during the training process.

[0203] Step S29, when the generator output of the updated generative adversarial model is the target performance indicator, the updating is ended to obtain the updated generative adversarial model.

[0204] It should be noted that the updated generative adversarial model refers to a model in which the parameters of the generator and discriminator have been updated according to the loss function and optimizer after a series of training iterations. The updated model should have better performance, that is, the generator can generate samples that are closer to real data, and the discriminator can distinguish between real and fake data more accurately. The target performance indicator is a pre-set performance standard used to evaluate whether the quality of the generator output meets specific requirements. The target performance indicator provides a clear goal for model training. When the generator output meets or exceeds this standard, the training can be considered successful.

[0205] It is understandable that when the generator output meets the target performance indicator, further training iterations are stopped and the current model is put into production use.

[0206] This embodiment provides a method for generating vehicle battery optimization suggestions based on a generative adversarial model, which simulates battery performance parameters by training a generator and uses a discriminator to evaluate and guide the output of the generator, thereby achieving accurate prediction and optimization of battery performance. This process involves initializing the weights of the generator and the discriminator, calculating the loss function and its gradient, and using the Adam optimizer to update the weights according to these gradients. Through continuous iterations, the generator learns to generate more realistic battery performance data, while the discriminator improves its ability to distinguish true and false data. When the generator output reaches the preset performance index, that is, it can generate sufficiently realistic battery performance indicators, the training is terminated and the final optimized generative adversarial model is obtained. It not only improves the battery utilization efficiency, but also enhances the intelligent level of battery management, provides strong technical support for the practical application of new energy electric vehicles, and achieves the beneficial effects of improving battery performance and extending battery life.

[0207] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the method for generating vehicle battery optimization recommendations based on a generative adversarial model in the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0208] This application also provides a vehicle battery optimization suggestion generation device based on a generative adversarial model, please refer to Figure 3 , the vehicle battery optimization suggestion generating device based on the generative adversarial model comprises:

[0209] The acquisition module 10 is used to acquire the performance parameters of the vehicle uploaded by the vehicle end;

[0210] The processing module 20 is used to pre-process the performance parameters to obtain pre-processed performance parameters;

[0211] An obtaining module 30 is used to obtain performance parameter weights and target performance parameters according to the preprocessing performance parameters and the generative adversarial model;

[0212] The optimization module 40 is used to obtain optimization suggestions based on the target performance parameters and the performance parameter weights, and send the optimization suggestions to the vehicle end.

[0213] The vehicle battery optimization suggestion generation device based on the generative adversarial model provided by the present application adopts the vehicle battery optimization suggestion generation method based on the generative adversarial model in the above-mentioned embodiment, which can solve the technical problem of how to optimize the energy efficiency of new energy electric vehicles in multiple dimensions. Compared with the prior art, the beneficial effects of the vehicle battery optimization suggestion generation device based on the generative adversarial model provided by the present application are the same as the beneficial effects of the vehicle battery optimization suggestion generation method based on the generative adversarial model provided by the above-mentioned embodiment, and the other technical features of the vehicle battery optimization suggestion generation device based on the generative adversarial model are the same as the features disclosed in the above-mentioned embodiment method, which will not be repeated here.

[0214] In one embodiment, the optimization module 40 is also used to obtain an optimization index threshold of the vehicle; obtain a target optimization index based on the target performance parameter and the performance parameter weight; when the target optimization index is greater than the optimization index threshold, obtain a speed adjustment suggestion, an energy recovery level adjustment suggestion and an acceleration adjustment suggestion based on the target optimization index; obtain a comprehensive optimization suggestion based on the speed adjustment suggestion, the energy recovery level adjustment suggestion and the acceleration adjustment suggestion; and convert the comprehensive optimization suggestion into a format to obtain an optimization suggestion.

[0215] In one embodiment, the optimization module 40 is further used to obtain the current speed, speed threshold, current energy recovery efficiency, energy recovery efficiency threshold, current energy recovery level, preset energy level, current acceleration and acceleration threshold of the vehicle when the target optimization index is greater than the optimization index threshold; when the current speed is greater than the speed threshold, the target speed is obtained according to the current speed and the target optimization index, and a speed adjustment suggestion is obtained according to the target speed; when the current energy recovery efficiency is less than the energy recovery efficiency threshold and the current energy recovery level is less than the preset energy level, the target recovery level is obtained according to the current energy recovery level and the target optimization index, and an energy recovery level adjustment suggestion is obtained according to the target recovery level; when the current acceleration is greater than the acceleration threshold, the target acceleration is obtained according to the current acceleration and the target optimization index, and an acceleration adjustment suggestion is obtained according to the target acceleration.

[0216] In one embodiment, the processing module 20 is also used to identify missing fields of the performance parameters, and fill the missing fields according to the average value of the vehicle's historical data or the average value of the missing fields to obtain complete performance parameters; obtain a first average value and a first standard deviation of the complete performance parameters and a second average value and a second standard deviation of the vehicle set parameters; obtain standard performance parameters based on at least one of the complete performance parameters, the first average value, the first standard deviation, the second average value and the second standard deviation; sort the standard performance parameters according to the upload time, and divide the sorted standard performance parameters according to preset time windows to obtain preprocessing performance parameters.

[0217] In one embodiment, the processing module 20 is also used to obtain a normal value range of a complete performance parameter based on the first mean value and the first standard deviation; obtain a normal performance parameter based on the complete performance parameter and the normal value range; obtain a local standardized parameter based on the normal performance parameter, the first mean value and the first standard deviation; obtain a global standardized parameter based on the local standardized parameter, the second mean value and the second standard deviation, and obtain a standard performance parameter based on the global standardized parameter.

[0218] In one embodiment, the obtaining module 30 is further used to obtain the initial performance parameter weight, the generator of the adversarial model and the discriminator of the adversarial model;

[0219] According to the generator and the discriminator, a generator loss function and a discriminator loss function are obtained; according to the generator loss function, the initial performance parameter weights are optimized to obtain target performance parameter weights; according to the generator loss function, the discriminator loss function and a preset optimizer, the generative adversarial model is updated; when the generator output of the updated generative adversarial model is the target performance indicator, the update is terminated to obtain an updated generative adversarial model.

[0220] In one embodiment, the obtaining module 30 is also used to obtain a weight parameter gradient based on the generator loss function and the initial performance parameter weight; obtain an optimized performance parameter weight based on the initial performance parameter weight, the weight parameter gradient and a preset learning rate; and obtain a target performance parameter weight based on the optimized performance parameter weight when the weight parameter gradient is equal to a preset gradient value or the generator loss function is a preset output.

[0221] The present application provides a device for generating vehicle battery optimization suggestions based on a generative adversarial model. The device for generating vehicle battery optimization suggestions based on a generative adversarial model includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for generating vehicle battery optimization suggestions based on a generative adversarial model in the above-mentioned embodiment one.

[0222] Reference below Figure 4 , which shows a schematic diagram of the structure of a vehicle battery optimization suggestion generation device based on a generative adversarial model suitable for implementing an embodiment of the present application. The vehicle battery optimization suggestion generation device based on a generative adversarial model in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The vehicle battery optimization suggestion generation device based on the generative adversarial model shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0223] like Figure 4As shown, the vehicle battery optimization suggestion generating device based on the generative adversarial model may include a processing device 1001 (such as a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM: Read Only Memory) 1002 or the program loaded from the storage device 1003 to the random access memory (RAM: Random Access Memory) 1004. Various programs and data required for the operation of the vehicle battery optimization suggestion generating device based on the generative adversarial model are also stored in RAM1004. The processing device 1001, ROM1002 and RAM1004 are connected to each other through a bus 1005. The input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 can allow the vehicle battery optimization suggestion generation device based on the generative adversarial model to communicate wirelessly or wired with other devices to exchange data. Although the figure shows a vehicle battery optimization suggestion generation device based on the generative adversarial model with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have instead.

[0224] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0225] The vehicle battery optimization suggestion generation device based on the generative adversarial model provided by the present application adopts the vehicle battery optimization suggestion generation method based on the generative adversarial model in the above-mentioned embodiment, which can solve the technical problem of how to optimize the energy efficiency of new energy electric vehicles in multiple dimensions. Compared with the prior art, the beneficial effects of the vehicle battery optimization suggestion generation device based on the generative adversarial model provided by the present application are the same as the beneficial effects of the vehicle battery optimization suggestion generation method based on the generative adversarial model provided by the above-mentioned embodiment, and the other technical features in the vehicle battery optimization suggestion generation device based on the generative adversarial model are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0226] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0227] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0228] The present application provides a computer-readable storage medium having computer-readable program instructions (i.e., computer programs) stored thereon, and the computer-readable program instructions are used to execute the vehicle battery optimization recommendation generation method based on the generative adversarial model in the above-mentioned embodiment.

[0229] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0230] The above-mentioned computer-readable storage medium may be included in the vehicle battery optimization suggestion generation device based on the generative adversarial model; or it may exist independently without being assembled into the vehicle battery optimization suggestion generation device based on the generative adversarial model.

[0231] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by a vehicle battery optimization suggestion generating device based on a generative adversarial model, the vehicle battery optimization suggestion generating device based on a generative adversarial model: obtains the performance parameters of the vehicle uploaded by the vehicle end;

[0232] The performance parameters are preprocessed to obtain preprocessed performance parameters; performance parameter weights and target performance parameters are obtained based on the preprocessed performance parameters and a generative adversarial model; optimization suggestions are obtained based on the target performance parameters and the performance parameter weights, and the optimization suggestions are sent to the vehicle end.

[0233] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0234] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0235] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0236] The readable storage medium provided by the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned vehicle battery optimization suggestion generation method based on the generative adversarial model, and can solve the technical problem of how to optimize the energy efficiency of new energy electric vehicles in multiple dimensions. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as the beneficial effects of the vehicle battery optimization suggestion generation method based on the generative adversarial model provided by the above-mentioned embodiment, and will not be repeated here.

[0237] The present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for generating vehicle battery optimization recommendations based on a generative adversarial model.

[0238] The computer program product provided by this application can solve the technical problem of how to optimize the energy efficiency of new energy electric vehicles in multiple dimensions. Compared with the prior art, the beneficial effects of the computer program product provided by this application are the same as the beneficial effects of the vehicle battery optimization suggestion generation method based on the generative adversarial model provided in the above embodiment, which will not be repeated here.

[0239] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A method for generating vehicle battery optimization suggestions based on a generative adversarial model, characterized in that: The method comprises: Obtain the vehicle performance parameters uploaded by the vehicle end; Preprocessing the performance parameters to obtain preprocessed performance parameters; Obtaining performance parameter weights and target performance parameters according to the preprocessing performance parameters and the generative adversarial model; An optimization suggestion is obtained according to the target performance parameter and the performance parameter weight, and the optimization suggestion is sent to the vehicle end.

2. The method according to claim 1, characterized in that The step of obtaining optimization suggestions according to the target performance parameter and the performance parameter weight comprises: Get the optimization index threshold of the vehicle; Obtaining a target optimization index according to the target performance parameter and the performance parameter weight; When the target optimization index is greater than the optimization index threshold, obtaining a speed adjustment suggestion, an energy recovery level adjustment suggestion, and an acceleration adjustment suggestion according to the target optimization index; Obtaining a comprehensive optimization suggestion according to the speed adjustment suggestion, the energy recovery level adjustment suggestion, and the acceleration adjustment suggestion; The comprehensive optimization suggestion is formatted to obtain an optimization suggestion.

3. The method according to claim 2, characterized in that When the target optimization index is greater than the optimization index threshold, the step of obtaining a speed adjustment suggestion, an energy recovery level adjustment suggestion, and an acceleration adjustment suggestion according to the target optimization index comprises: When the target optimization index is greater than the optimization index threshold, obtaining the current speed, speed threshold, current energy recovery efficiency, energy recovery efficiency threshold, current energy recovery level, preset energy level, current acceleration and acceleration threshold of the vehicle; When the current speed is greater than the speed threshold, a target speed is obtained according to the current speed and the target optimization index, and a speed adjustment suggestion is obtained according to the target speed; When the current energy recovery efficiency is less than the energy recovery efficiency threshold and the current energy recovery level is less than the preset energy level, a target recovery level is obtained according to the current energy recovery level and the target optimization index, and an energy recovery level adjustment suggestion is obtained according to the target recovery level; When the current acceleration is greater than the acceleration threshold, a target acceleration is obtained according to the current acceleration and the target optimization index, and an acceleration adjustment suggestion is obtained according to the target acceleration.

4. The method according to claim 1, characterized in that The step of preprocessing the performance parameters to obtain preprocessed performance parameters comprises: Identify missing fields of the performance parameters, and fill the missing fields according to an average value of historical data of the vehicle or an average value of the missing fields to obtain complete performance parameters; Obtaining a first mean value and a first standard deviation of the complete performance parameter and a second mean value and a second standard deviation of the vehicle set parameter; Obtaining a standard performance parameter according to at least one of the complete performance parameter, the first average value, the first standard deviation, the second average value, and the second standard deviation; The standard performance parameters are sorted according to the upload time, and the sorted standard performance parameters are divided according to the preset time window to obtain the preprocessing performance parameters.

5. The method according to claim 4, characterized in that The step of obtaining a standard performance parameter according to at least one of the complete performance parameter, the first average value, the first standard deviation, the second average value, and the second standard deviation comprises: Obtaining a normal value range of a complete performance parameter according to the first mean value and the first standard deviation; Obtaining normal performance parameters according to the complete performance parameters and the normal value range; Obtaining a local standardized parameter according to the normal performance parameter, the first average value, and the first standard deviation; A global standardized parameter is obtained according to the local standardized parameter, the second average value and the second standard deviation, and a standard performance parameter is obtained according to the global standardized parameter.

6. The method according to claim 1, characterized in that Before the step of obtaining the performance parameter weights and the target performance parameters according to the preprocessing performance parameters and the generative adversarial model, the step further includes: Obtain initial performance parameter weights, the generator of the adversarial model, and the discriminator of the adversarial model; According to the generator and the discriminator, a generator loss function and a discriminator loss function are obtained; Optimizing the initial performance parameter weights according to the generator loss function to obtain target performance parameter weights; Update the generative adversarial model according to the generator loss function, the discriminator loss function and the preset optimizer; When the generator output of the updated generative adversarial model is the target performance indicator, the updating is terminated to obtain the updated generative adversarial model.

7. The method according to claim 6, characterized in that The step of optimizing the initial performance parameter weight according to the generator loss function to obtain the target performance parameter weight comprises: Obtaining a weight parameter gradient according to the generator loss function and the initial performance parameter weight; Obtaining an optimized performance parameter weight according to the initial performance parameter weight, the weight parameter gradient and the preset learning rate; When the weight parameter gradient is equal to a preset gradient value or the generator loss function is a preset output, the target performance parameter weight is obtained according to the optimized performance parameter weight.

8. A vehicle battery optimization suggestion generation device based on a generative adversarial model, characterized in that: The device comprises: The acquisition module is used to obtain the performance parameters of the vehicle uploaded by the vehicle end; A processing module, used for preprocessing the performance parameters to obtain preprocessed performance parameters; An obtaining module, used for obtaining performance parameter weights and target performance parameters according to the preprocessing performance parameters and the generative adversarial model; The optimization module is used to obtain optimization suggestions based on the target performance parameters and the performance parameter weights, and send the optimization suggestions to the vehicle end.

9. A vehicle battery optimization suggestion generation device based on a generative adversarial model, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the method for generating vehicle battery optimization recommendations based on a generative adversarial model as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for generating vehicle battery optimization recommendations based on a generative adversarial model as described in any one of claims 1 to 7 are implemented.

Images