Energy storage and power generation status assessment method and system for low-speed electric vehicle batteries

Abstract

The present invention relates to the technical field of energy storage and power generation status assessment, specifically to an energy storage and power generation status assessment method and system for low-speed electric vehicle batteries. The system includes: a charging and discharging behavior analysis module, which establishes a vehicle charging and discharging time point matrix based on the collected low-speed electric vehicle charging and discharging history data, and the matrix elements represent the charging and discharging behavior of the target vehicle at the target time point. By calculating the similarity between the vehicles, a group of vehicles with the same charging and discharging behavior as the target vehicle is captured. In the present invention, by collecting the low-speed electric vehicle charging and discharging history data, a vehicle charging and discharging time point matrix is ​​established, and the vehicle group and charging and discharging time pattern with similar charging and discharging behavior are captured, thereby achieving accurate classification and analysis of the vehicle charging and discharging behavior. The cosine similarity is used to analyze the charging and discharging behavior and time pattern, and a recommended charging and discharging time and frequency strategy is generated, so that the charging and discharging plan is more scientific and reasonable.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage and power generation status assessment, and in particular to an energy storage and power generation status assessment method and system for a low-speed electric vehicle battery. Background Art

[0002] Energy storage power generation status assessment mainly involves monitoring, analyzing and evaluating the working status and performance of power energy storage equipment and systems. This technology uses various sensors and data acquisition devices to obtain parameters such as voltage, current, temperature, etc. of batteries or other energy storage devices in real time, and combines intelligent algorithms and data processing technology to conduct a comprehensive assessment of the health status, charging and discharging efficiency, and energy conversion efficiency of the energy storage system. The assessment results can be used to predict the service life of the equipment, optimize the operation strategy of the energy storage system, improve energy utilization efficiency, and ultimately ensure the safe and reliable operation of the energy storage system.

[0003] The energy storage and power generation status assessment method and system for low-speed electric vehicle batteries are mainly used to assess the working status and energy storage and power generation performance of low-speed electric vehicle batteries. Its purpose is to monitor and analyze the charge and discharge status, health status and energy storage efficiency of low-speed electric vehicle batteries in real time through specific assessment methods and systems. Through this system, potential battery problems can be discovered in a timely manner, the battery use and maintenance strategy can be optimized, the battery life can be extended, and the endurance and overall performance of low-speed electric vehicles can be improved.

[0004] However, in the process of evaluating the energy storage and power generation status, the existing technology mainly relies on sensors and data acquisition devices to obtain the basic parameters of the battery, and combines intelligent algorithms for comprehensive evaluation, lacking detailed analysis of charging and discharging behaviors and accurate recommendations for optimization strategies. Since it only relies on the monitoring of basic parameters and a single algorithm analysis, it is difficult to capture the complex differences and time patterns of charging and discharging behaviors between vehicles, resulting in a relatively extensive formulation of charging and discharging strategies that cannot fully take into account the characteristics and needs of individual vehicles. The existing technology also has deficiencies in strategy verification and dynamic adjustment, lacks real-time monitoring and feedback mechanisms for the effectiveness of strategy implementation, and has poor adaptability and long-term effectiveness, which may affect the long-term use of the battery and the overall performance of the vehicle. These deficiencies may lead to the failure to promptly discover and handle potential battery problems in actual operations, and the inability to effectively guarantee battery life and performance, which in turn affects the endurance and user experience of low-speed electric vehicles. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a method and system for evaluating the energy storage and power generation status of a low-speed electric vehicle battery.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: the energy storage and power generation status assessment system for low-speed electric vehicle batteries comprises:

[0007] The charging and discharging behavior analysis module establishes a vehicle charging and discharging time point matrix based on the collected low-speed electric vehicle charging and discharging history data. The matrix elements represent the charging and discharging behavior of the target vehicle at the target time point. By calculating the similarity between the vehicles, the vehicle group with the same charging and discharging behavior as the target vehicle is captured. By calculating the similarity between the charging and discharging time points, the same charging and discharging time pattern is captured and the same charging and discharging behavior is generated.

[0008] The collaborative filtering analysis module calculates the similarity between each vehicle based on the similar charging and discharging behaviors, analyzes the similarity of the charging and discharging behaviors by cosine similarity, obtains a group of vehicles with the same charging and discharging behaviors as the target vehicle, calculates the similarity between multiple charging and discharging time points, analyzes the charging and discharging time patterns by similarity, and obtains the charging and discharging similarity analysis results;

[0009] An optimization strategy recommendation module generates a recommended charging and discharging time and frequency strategy based on the charging and discharging similarity analysis results and the charging and discharging behaviors of similar vehicles. It uses the scoring prediction technology to score multiple charging and discharging time points of the target vehicle, selects the strategy with the best score as the recommended strategy, and generates the optimal charging and discharging strategy.

[0010] The model verification module applies the optimal charging and discharging strategy to the vehicle, monitors the performance and life of the vehicle battery, compares the battery status data before and after the strategy is applied, compares the charging amount, discharging amount and service life parameters, evaluates the strategy effect, and obtains the charging and discharging strategy effect;

[0011] The strategy adjustment module calls new charging and discharging data based on the effect of the charging and discharging strategy, updates the vehicle charging and discharging time point matrix, recalculates the similarities between vehicles and between time points, adjusts the collaborative filtering model, updates the recommendation strategy through the new similarity analysis results, and generates an optimized charging and discharging strategy.

[0012] As a further solution of the present invention, the same type of charging and discharging behaviors include charging frequency similarity, discharge time similarity and charging duration similarity, the charging and discharging similarity analysis results include a vehicle-to-vehicle similarity matrix, a time point similarity matrix and a similarity score list, the optimal charging and discharging strategy includes a recommended charging time, a recommended discharge time and a recommended charging frequency, the charging and discharging strategy effects include changes in battery charge amount, changes in battery discharge amount and changes in battery service life, and the optimized charging and discharging strategy includes an adjusted charging time, an adjusted discharge time and an updated charging frequency.

[0013] As a further solution of the present invention, the charging and discharging behavior analysis module includes:

[0014] The data collection submodule performs data cleaning and preprocessing based on the collected historical data of low-speed electric vehicle charging and discharging, removes noise and outliers, and establishes a vehicle charging and discharging time point matrix. The matrix elements represent the charging and discharging behavior of the target vehicle at the target time point, and generates a charging and discharging time point matrix;

[0015] A similarity calculation submodule performs data standardization processing based on the charge and discharge time point matrix, calculates the similarity of the charge and discharge behaviors between vehicles, identifies vehicles with similar charge and discharge behaviors, captures a group of vehicles with similar charge and discharge behaviors as the target vehicle, and generates a group of similar vehicles;

[0016] The behavior pattern generation submodule performs cluster analysis on the charging and discharging time point data based on the same type of vehicle group, identifies the charging and discharging time pattern, captures the same type of charging and discharging time pattern, and generates the same type of charging and discharging behavior.

[0017] As a further solution of the present invention, the collaborative filtering analysis module includes:

[0018] A similarity calculation submodule calculates the similarity between each vehicle based on the same type of charging and discharging behavior of each vehicle, analyzes the similarity of charging and discharging behavior through cosine similarity, uses the charging and discharging data of the vehicle to compare the charging start time and the end time, calculates the time difference between charging and discharging, and generates a vehicle similarity matrix;

[0019] The charging and discharging behavior analysis submodule obtains a vehicle group with the same charging and discharging behavior as the target vehicle based on the vehicle similarity matrix, performs cluster analysis on the charging and discharging behavior of the vehicle group, and uses the charging frequency and duration of multiple vehicles in the group to perform clustering of similar features to obtain the characteristics of the vehicle group;

[0020] The charging and discharging time pattern analysis submodule calculates the similarity between multiple charging and discharging time points based on the characteristics of the same type of vehicle group, analyzes the charging and discharging time pattern through similarity, uses time series comparison to identify the charging peak and valley time periods, and generates charging and discharging similarity analysis results.

[0021] As a further solution of the present invention, the optimization strategy recommendation module includes:

[0022] A strategy generation submodule, based on the charge and discharge similarity analysis results and combined with the charge and discharge behaviors of similar vehicles, collects and analyzes data to perform statistics and optimization on the charge and discharge time and frequency, and generates a recommended charge and discharge strategy;

[0023] The scoring prediction submodule scores multiple charging and discharging time points of the target vehicle based on the recommended charging and discharging strategy, calculates the scoring value of each time point through data analysis and prediction, and generates a charging and discharging scoring result;

[0024] The strategy selection submodule selects the best-scoring charge and discharge strategy based on the charge and discharge scoring results, determines the best solution through comparison and selection, and uses it as the final recommended strategy to generate the best charge and discharge strategy.

[0025] As a further solution of the present invention, the model verification module includes:

[0026] A strategy application submodule, based on the optimal charging and discharging strategy, applies the strategy to the vehicle, monitors the vehicle status during the strategy execution, records the time points of charging and discharging, uses an on-board data recorder to record the charging voltage and current, monitors the start and end time of each charging and discharging, and generates a strategy application record;

[0027] The battery monitoring submodule monitors the performance and life of the vehicle battery based on the strategy application record, uses sensors to collect the voltage, current and temperature data of the battery, and records the change in battery capacity during each charge and discharge, regularly checks the internal resistance and voltage stability of the battery, and generates battery performance data;

[0028] The effect evaluation submodule compares the battery status before and after the application of the strategy based on the battery performance data, calculates the charging capacity, discharging capacity and service life parameters, uses the data comparison method to judge the strategy effect, compares the battery charging and discharging efficiency and life changes in multiple time periods, and obtains the charging and discharging strategy effect.

[0029] As a further solution of the present invention, the policy adjustment module includes:

[0030] The data update submodule calls new charge and discharge data based on the effect of the charge and discharge strategy, performs data collection and preprocessing, removes noise and outliers, updates the vehicle charge and discharge time point matrix, and generates an updated charge and discharge matrix;

[0031] A similarity recalculation submodule, based on the updated charge-discharge matrix, recalculates the similarity between vehicles and the similarity between time points, and generates a new similarity result through data standardization processing and similarity calculation;

[0032] The strategy optimization submodule adjusts the collaborative filtering model based on the new similarity result, updates the recommendation strategy, and generates an optimized charging and discharging strategy through strategy optimization and adjustment.

[0033] The energy storage and power generation status evaluation method for a low-speed electric vehicle battery is performed based on the energy storage and power generation status evaluation system for a low-speed electric vehicle battery, and comprises the following steps:

[0034] S1: Based on the collected historical data of low-speed electric vehicle charging and discharging, a matrix of vehicle charging and discharging time points is established by using a matrix construction method. The similarity between vehicles is calculated to capture the vehicle group with the same charging and discharging behavior as the target vehicle. Then, the similarity between charging and discharging time points is calculated to capture the same charging and discharging time patterns and generate the same charging and discharging behavior.

[0035] S2: Based on the similar charging and discharging behaviors, a cosine similarity analysis method is used to calculate the similarity between each vehicle, and the similarity of the charging and discharging behaviors is analyzed through the similarity to obtain a group of vehicles with the same charging and discharging behaviors as the target vehicle, and the similarity between multiple charging and discharging time points is calculated. The charging and discharging time pattern is analyzed through the similarity to obtain a charging and discharging similarity analysis result;

[0036] S3: Based on the charge and discharge similarity analysis result, a scoring prediction method is used to combine the charge and discharge behaviors of similar vehicles to generate a recommended charge and discharge time and frequency strategy, score multiple charge and discharge time points of the target vehicle, select the best strategy as the recommended strategy through scoring, and generate the best charge and discharge strategy;

[0037] S4: Based on the optimal charging and discharging strategy, monitor the changes in vehicle battery performance and life, use data comparison and analysis methods to compare the battery status data before and after the application of the strategy, compare the charging amount, discharging amount and service life parameters, evaluate the strategy effect, obtain the charging and discharging strategy effect, call new charging and discharging data, update the vehicle charging and discharging time point matrix, recalculate the similarity between vehicles and between time points, adjust the collaborative filtering model, update the recommended strategy through the new similarity analysis results, and generate an optimized charging and discharging strategy.

[0038] Compared with the prior art, the advantages and positive effects of the present invention are:

[0039] In the present invention, by collecting the historical data of low-speed electric vehicle charging and discharging, establishing a vehicle charging and discharging time point matrix, capturing vehicle groups and charging and discharging time patterns with similar charging and discharging behaviors, accurate classification and analysis of vehicle charging and discharging behaviors are achieved. The cosine similarity is used to analyze the charging and discharging behaviors and time patterns, and the recommended charging and discharging time and frequency strategies are generated to make the charging and discharging scheme more scientific and reasonable. The charging and discharging time points are scored through the scoring prediction technology, and the strategy with the best score is selected as the recommended strategy to ensure the optimality and effectiveness of the strategy. The recommended strategy is applied to the vehicle, the battery performance and life changes are monitored, and the charging and discharging strategy is verified and optimized, so that the battery life and performance are significantly improved. The charging and discharging data are continuously updated and the collaborative filtering model is adjusted to maintain the dynamic optimization of the strategy to ensure the long-term effectiveness and adaptability of the strategy. Compared with the existing technology, the above measures greatly improve the accuracy of battery health status assessment and charging and discharging strategy optimization, and improve the endurance and overall performance of low-speed electric vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0042] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0043] Embodiment 1

[0044] See also Figure 1 The present invention provides a technical solution: an energy storage and power generation status assessment system for a low-speed electric vehicle battery comprises:

[0045] The charging and discharging behavior analysis module establishes a vehicle charging and discharging time point matrix based on the collected low-speed electric vehicle charging and discharging history data. The matrix elements represent the charging and discharging behavior of the target vehicle at the target time point. By calculating the similarity between the vehicles, the vehicle group with the same charging and discharging behavior as the target vehicle is captured. By calculating the similarity between the charging and discharging time points, the same charging and discharging time pattern is captured and the same charging and discharging behavior is generated.

[0046] The collaborative filtering analysis module calculates the similarity between each vehicle based on similar charging and discharging behaviors, analyzes the similarity of charging and discharging behaviors through cosine similarity, obtains the vehicle group with similar charging and discharging behaviors as the target vehicle, calculates the similarity between multiple charging and discharging time points, analyzes the charging and discharging time patterns through similarity, and obtains the charging and discharging similarity analysis results;

[0047] The optimization strategy recommendation module generates recommended charging and discharging time and frequency strategies based on the charging and discharging similarity analysis results and the charging and discharging behaviors of similar vehicles. It uses the scoring prediction technology to score multiple charging and discharging time points of the target vehicle, selects the strategy with the best score as the recommended strategy, and generates the optimal charging and discharging strategy.

[0048] The model verification module applies the optimal charging and discharging strategy to the vehicle, monitors the vehicle battery performance and life changes, compares the battery status data before and after the strategy is applied, compares the charging amount, discharging amount and service life parameters, evaluates the strategy effect, and obtains the charging and discharging strategy effect;

[0049] The strategy adjustment module calls new charging and discharging data based on the effect of the charging and discharging strategy, updates the vehicle charging and discharging time point matrix, recalculates the similarities between vehicles and between time points, adjusts the collaborative filtering model, updates the recommended strategy through the new similarity analysis results, and generates an optimized charging and discharging strategy.

[0050] Similar charging and discharging behaviors include charging frequency similarity, discharge time similarity and charging duration similarity. The charging and discharging similarity analysis results include the vehicle-to-vehicle similarity matrix, the time point similarity matrix and the similarity score list. The optimal charging and discharging strategy includes the recommended charging time, the recommended discharge time and the recommended charging frequency. The charging and discharging strategy effects include the change in battery charge amount, the change in battery discharge amount and the change in battery service life. The optimized charging and discharging strategy includes the adjusted charging time, the adjusted discharge time and the updated charging frequency.

[0051] In the charging and discharging behavior analysis module, by collecting the historical data of low-speed electric vehicle charging and discharging, the time series analysis method is used to establish the vehicle charging and discharging time point matrix. The matrix elements represent the charging and discharging behavior of the target vehicle at a specific time point. The K-means clustering algorithm is used to cluster the time series data according to the charging and discharging behavior, and the Euclidean distance between each vehicle and other vehicles is calculated to capture the vehicle group with similar charging and discharging behavior to the target vehicle. At the same time, the dynamic time warping (DTW) algorithm is used to analyze the similarity between different time points, capture similar charging and discharging time patterns, and generate similar charging and discharging behaviors. The data generated by this module includes the charging and discharging time point matrix and the similarity matrix. The data can be used to identify the charging and discharging mode of the vehicle, providing basic data support for subsequent modules.

[0052] In the collaborative filtering analysis module, the similarity between each vehicle is calculated based on the data of similar charging and discharging behaviors. The cosine similarity analysis method is used to calculate the similarity of the charging and discharging behaviors of each vehicle, and obtain a group of vehicles with similar charging and discharging behaviors to the target vehicle. The specific operation is to first vectorize the charging and discharging behavior data of each vehicle, and then calculate the cosine similarity between each pair of vehicles to obtain a vehicle similarity matrix. Then, through the clustering algorithm, similar charging and discharging behaviors are classified to generate a similarity matrix between multiple charging and discharging time points, and the charging and discharging time pattern is analyzed by similarity to obtain the charging and discharging similarity analysis results. The output of this module includes the similarity matrix between vehicles and the time point similarity matrix, which provides a data basis for the optimization strategy recommendation module.

[0053] In the optimization strategy recommendation module, based on the results of the charge and discharge similarity analysis and combined with the charge and discharge behaviors of similar vehicles, the recommended charge and discharge time and frequency strategies are generated, and the score prediction technology is used to score the multiple charge and discharge time points of the target vehicle. The specific operation is to use the collaborative filtering recommendation algorithm, take the charge and discharge time points of each vehicle as the score matrix, calculate the predicted score, and select the strategy with the best score as the recommended strategy based on the score results. The generated file includes the optimal charge and discharge strategy, specifically the recommended charging time, recommended discharge time, and recommended charging frequency. The effect of this module is to generate personalized charge and discharge strategies for each vehicle, improve battery efficiency and extend battery life.

[0054] In the model verification module, the optimal charge and discharge strategy is applied to the vehicle to monitor the changes in vehicle battery performance and life. The specific operation is to collect battery status data before and after the application of the strategy, and calculate and compare the changes in charge, discharge and service life parameters. Regression analysis and statistical test methods are used to compare the data differences before and after application to judge the effect of the strategy. The generated file includes a charge and discharge strategy effect evaluation report, which includes changes in battery charge, battery discharge and battery service life. The role of this module is to verify the effectiveness of the recommended strategy and provide a basis for subsequent strategy adjustments.

[0055] In the strategy adjustment module, based on the effect of the charging and discharging strategy, new charging and discharging data is called, the vehicle charging and discharging time point matrix is ​​updated, the similarity between vehicles and time points is recalculated, and the collaborative filtering model is adjusted using an adaptive collaborative filtering algorithm. The specific operation is to input the new charging and discharging data into the model, update the charging and discharging time point matrix, use the new data to recalculate the similarity between vehicles and time points, and update the recommended strategy through the new similarity analysis results. The generated file includes the optimized charging and discharging strategy, specifically the adjusted charging time, adjusted discharging time, and updated charging frequency. The effect of this module is to continuously optimize the charging and discharging strategy based on the latest data to ensure the continued effectiveness of the recommended strategy.

[0056] In this embodiment, the charge and discharge behavior analysis module includes:

[0057] The data collection submodule performs data cleaning and preprocessing based on the collected historical data of low-speed electric vehicle charging and discharging, removes noise and outliers, and establishes a vehicle charging and discharging time point matrix. The matrix elements represent the charging and discharging behavior of the target vehicle at the target time point, and generates a charging and discharging time point matrix;

[0058] The similarity calculation submodule performs data standardization based on the charge and discharge time point matrix, calculates the similarity of the charge and discharge behaviors between vehicles, identifies vehicles with similar charge and discharge behaviors, captures vehicle groups with similar charge and discharge behaviors as the target vehicle, and generates similar vehicle groups;

[0059] The behavior pattern generation submodule performs cluster analysis on the charging and discharging time point data based on similar vehicle groups, identifies the charging and discharging time patterns, captures similar charging and discharging time patterns, and generates similar charging and discharging behaviors.

[0060] In the data collection submodule, by collecting the historical charging and discharging data of low-speed electric vehicles, data cleaning and preprocessing techniques are used to remove noise and outliers in the data. Specific operations include using the Z-score method to detect and remove outliers, setting thresholds to mark data above or below a specific range as outliers and delete them. Next, linear interpolation is used to fill in missing values ​​to ensure data integrity. The processed data is organized in a time series format to establish a vehicle charging and discharging time point matrix. The rows of the matrix represent different vehicles, the columns represent specific time points, and the matrix elements represent the charging and discharging behavior of the vehicle at that time point, such as charging current, voltage and other values.

[0061] In the similarity calculation submodule, the data of the charge and discharge time point matrix is ​​normalized to eliminate the influence of numerical differences between different vehicles, and the Min-Max normalization method is used to scale the data to the same range. The specific operation is to calculate the minimum and maximum values ​​of each time point and convert each data point to a standardized value between 0 and 1. Then, the Euclidean distance is used to calculate the similarity of the charge and discharge behaviors between vehicles. The specific steps are to calculate the difference in charge and discharge behaviors between every two vehicles at each time point, sum them up and take the square root to get the Euclidean distance value. Next, the density clustering algorithm (DBSCAN) is used to identify vehicle groups with similar charge and discharge behaviors. By setting the distance threshold and the minimum group size, the vehicle groups with similar charge and discharge behaviors to the target vehicle are screened out. Generate a file of similar vehicle groups, which contains each target vehicle and its corresponding similar vehicle list, to provide data support for subsequent behavior pattern analysis.

[0062] In the behavior pattern generation submodule, cluster analysis is performed on the charging and discharging time point data based on the data of the same vehicle group to identify the charging and discharging time pattern. The K-means clustering algorithm is used. The specific steps are to first determine the cluster number k value, use the elbow method to select the optimal k value, and then randomly initialize k cluster centers, assign each time point data to the nearest cluster center, and iteratively update the cluster center until convergence. Through cluster analysis, the same type of charging and discharging time pattern is captured, and the same type of charging and discharging behavior file is generated, which contains the charging and discharging time characteristics and behavior descriptions of each cluster. In this way, the charging and discharging rules of low-speed electric vehicles in different time periods can be identified, providing data support and providing a reference for further optimization strategy recommendation and evaluation.

[0063] In this embodiment, the collaborative filtering analysis module includes:

[0064] The similarity calculation submodule calculates the similarity between each vehicle based on the same charging and discharging behavior of each vehicle, analyzes the similarity of charging and discharging behavior through cosine similarity, uses the charging and discharging data of the vehicle to compare the charging start time and end time, calculates the time difference between charging and discharging, and generates a vehicle similarity matrix;

[0065] The charging and discharging behavior analysis submodule obtains the vehicle group with the same charging and discharging behavior as the target vehicle based on the vehicle similarity matrix. By clustering the charging and discharging behavior of the vehicle group, the charging frequency and duration of multiple vehicles in the group are used to cluster the same characteristics and obtain the characteristics of the same vehicle group.

[0066] The charging and discharging time pattern analysis submodule calculates the similarity between multiple charging and discharging time points based on the characteristics of the same type of vehicle group, analyzes the charging and discharging time patterns through similarity, uses time series comparison to identify the charging peak and valley time periods, and generates charging and discharging similarity analysis results.

[0067] In the similarity calculation submodule, the similarity between each vehicle is calculated based on the similar charging and discharging behavior of each vehicle, and the cosine similarity is used to analyze the similarity of the charging and discharging behavior. The specific steps are as follows: First, the charging and discharging data of each vehicle are collected, and the data includes the charging start time, charging end time, discharging start time and discharging end time. The data is preprocessed to remove outliers and noise data, and then the data of each vehicle is standardized so that it can be compared on a unified time scale. Next, the charging and discharging time difference of each vehicle at different time points is calculated to generate a time difference matrix. Using the cosine similarity formula, the similarity between each two vehicles in the time difference matrix is ​​calculated, and the results are stored in the vehicle similarity matrix. The cosine similarity formula is defined as the dot product of two vectors divided by the product of their modulus lengths. Through this formula, the similarity score between each vehicle is obtained. The generated vehicle similarity matrix can reflect the similarity of the charging and discharging behaviors between vehicles, providing a basis for subsequent analysis.

[0068] In the charging and discharging behavior analysis submodule, a group of vehicles with similar charging and discharging behaviors as the target vehicle is obtained based on the vehicle similarity matrix. The specific operation steps are as follows: First, select the target vehicle, and extract several vehicles with high similarity from the similarity matrix to form an initial group of similar vehicles. Next, cluster analysis is performed on the charging and discharging data of the vehicle group, and the K-means clustering algorithm is used to input the charging frequency and duration of the vehicle group as feature vectors. The K-means algorithm iteratively updates the cluster center and assigns samples to the nearest cluster center to ultimately determine the optimal clustering result. The clustering result obtained is the characteristics of the same vehicle group, which includes the charging frequency distribution and charging duration distribution of each vehicle in the group. This process accurately classifies vehicles with similar charging and discharging behaviors into one category through the cluster analysis method, providing strong support for the subsequent analysis of charging and discharging time patterns.

[0069] In the charging and discharging time pattern analysis submodule, the similarity between multiple charging and discharging time points is calculated based on the characteristics of the same type of vehicle group, and the charging and discharging time pattern is analyzed by similarity. The specific steps are as follows: First, the charging and discharging time data of the same type of vehicle group is sorted into time series data, which includes the charging and discharging records of each vehicle at different time points. Then, the dynamic time warping algorithm (Dynamic Time Warping, DTW) is used to calculate the similarity between different time series. The DTW algorithm finds the optimal alignment path between two time series by elastic matching of time series and calculates their similarity scores. Then, the similarity scores are analyzed to identify the charging peak and valley time periods, and generate the charging and discharging similarity analysis results. The results include the charging peak and valley information of each time period, which can help optimize the grid dispatch and resource allocation, and improve the charging and discharging efficiency. Through the detailed analysis of the charging and discharging time pattern, the behavior pattern of the vehicle group can be identified, providing data support, and then providing a reference for the planning and management of charging facilities.

[0070] In this embodiment, the optimization strategy recommendation module includes:

[0071] The strategy generation submodule collects and analyzes data based on the charge and discharge similarity analysis results and the charge and discharge behaviors of similar vehicles to generate a recommended charge and discharge strategy by collecting and analyzing the charge and discharge time and frequency statistics.

[0072] The scoring prediction submodule scores the target vehicle at multiple charging and discharging time points based on the recommended charging and discharging strategy. Through data analysis and prediction, it calculates the score value at each time point and generates the charging and discharging scoring results.

[0073] The strategy selection submodule selects the best-scoring charge and discharge strategy based on the charge and discharge scoring results, determines the best solution through comparison and selection, and uses it as the final recommended strategy to generate the optimal charge and discharge strategy.

[0074] In the strategy generation submodule, the charging and discharging time and frequency are statistically analyzed and optimized based on the charging and discharging similarity analysis results and the charging and discharging behaviors of similar vehicles. First, cluster analysis technology is used to group the charging and discharging data of similar vehicles according to time and frequency, and the K-means clustering algorithm is used to determine the optimal charging and discharging time period and frequency. Each cluster center represents the charging and discharging mode of a group of vehicles in a certain time period. Next, statistical analysis is performed to calculate the mean and variance of the charging time, discharging time, and charging frequency of the vehicles in each cluster, and identify the most common charging and discharging mode. Then, a genetic algorithm is used for parameter optimization, and the optimal charging and discharging strategy combination is found through crossover, mutation, and selection operations. This submodule generates a recommended charging and discharging strategy file, which contains the recommended charging time, discharging time, and charging frequency, providing basic data for the next step of score prediction.

[0075] In the score prediction submodule, multiple charging and discharging time points of the target vehicle are scored based on the recommended charging and discharging strategy, and the collaborative filtering algorithm is used for score prediction. The specific operation is to construct the charging and discharging time point data of each vehicle into a scoring matrix, and the matrix elements represent the charging and discharging behavior score of a vehicle at a specific time point. Then, using matrix decomposition techniques such as singular value decomposition (SVD), the scoring matrix is ​​decomposed into a user feature matrix and an item feature matrix to calculate the predicted score for each time point. In order to improve the prediction accuracy, regularization technology is also used to prevent the model from overfitting. The scoring results provide data support for subsequent strategy selection, ensuring that the selected charging and discharging strategy can maximize the performance and life of the battery.

[0076] In the strategy selection submodule, the best charging and discharging strategy is selected based on the charging and discharging scoring results, and the heuristic search algorithm is used for comparison and selection. The specific operation is to first sort all the recommended charging and discharging strategies by scoring, and select the strategies with the highest scores for detailed comparison. Then, a multi-objective optimization method is used to consider multiple factors such as battery life, charging efficiency, and user needs, and then the strategies are comprehensively evaluated to determine the optimal charging and discharging strategy. Finally, the optimal charging and discharging strategy file is generated, which contains detailed recommendations on charging time, discharging time, and charging frequency. Through the above content, it is ensured that the selected charging and discharging strategy is optimal in multiple dimensions, providing the most effective solution for battery management of low-speed electric vehicles. The effect of this module is to improve the efficiency of battery use, extend battery life, and provide users with specific operation guidance.

[0077] In this embodiment, the model verification module includes:

[0078] The strategy application submodule applies the strategy to the vehicle based on the optimal charging and discharging strategy, monitors the vehicle status during the strategy execution, records the charging and discharging time points, uses the on-board data recorder to record the charging voltage and current, monitors the start and end time of each charging and discharging, and generates strategy application records;

[0079] The battery monitoring submodule monitors the vehicle battery performance and life based on the strategy application records. It uses sensors to collect the battery voltage, current and temperature data, and records the battery capacity change after each charge and discharge. It regularly checks the battery internal resistance and voltage stability to generate battery performance data.

[0080] The effect evaluation submodule compares the battery status before and after the application of the strategy based on the battery performance data, calculates the charging capacity, discharging capacity and service life parameters, uses data comparison methods to judge the effect of the strategy, compares the battery charging and discharging efficiency and life changes in multiple time periods, and obtains the effect of the charging and discharging strategy.

[0081] In the strategy application submodule, the strategy is applied to the vehicle based on the optimal charging and discharging strategy. The specific operations include monitoring the vehicle status during the strategy execution process, recording the time points of charging and discharging, using the on-board data recorder to record the charging voltage and current, monitoring the start and end time of each charging and discharging, and generating strategy application records. The on-board data recorder stores the recorded charging voltage and current data in real time and transmits it to the central database via a wireless network. The data includes the time point, voltage, current value and duration of each charging and discharging. The generated strategy application record contains a detailed record of the key parameters of the vehicle during the strategy execution process, providing data support for subsequent battery monitoring and effect evaluation.

[0082] In the battery monitoring submodule, the vehicle battery performance and life are monitored through strategy-based application records. The specific operations include using sensors to collect the battery voltage, current and temperature data, and recording the battery capacity changes for each charge and discharge. Sensors are installed at various key locations of the battery pack to monitor the working status of the battery in real time and transmit the collected data to the data processing unit. The data processing unit pre-processes the voltage, current and temperature data, removes noise and outliers, and then calculates the change in battery capacity. In addition, the internal resistance and voltage stability of the battery are checked regularly, and the internal resistance changes and voltage fluctuations of the battery are measured by professional instruments to generate battery performance data. The generated battery performance data includes the battery capacity change curve, temperature change trend and internal resistance fluctuation. The data is used to evaluate the health status of the battery and predict its service life.

[0083] In the effect evaluation submodule, the battery status before and after the strategy is applied is compared based on the battery performance data. The specific operations include calculating the charge, discharge and service life parameters. Using the data comparison method, the battery performance data before and after the strategy is applied is first classified and sorted to form a comparison data set. Then, the statistical analysis method is used to compare the charging efficiency and discharge efficiency of the battery in different time periods, and the average charging time, average discharge time and energy conversion efficiency are calculated. In addition, the life prediction algorithm is used to analyze the changes in the battery life under different strategies, and a battery life evaluation report is generated. The evaluation report includes a comparison chart of the battery charging and discharging efficiency before and after the strategy is applied, a life change curve and related statistical parameters.

[0084] In this embodiment, the policy adjustment module includes:

[0085] The data update submodule calls new charging and discharging data based on the charging and discharging strategy effect, performs data collection and preprocessing, removes noise and outliers, updates the vehicle charging and discharging time point matrix, and generates an updated charging and discharging matrix;

[0086] The similarity recalculation submodule recalculates the similarity between vehicles and the similarity between time points based on the updated charge and discharge matrix, and generates new similarity results through data standardization and similarity calculation;

[0087] The strategy optimization submodule adjusts the collaborative filtering model based on the new similarity results, updates the recommendation strategy, and generates an optimized charging and discharging strategy through strategy optimization and adjustment.

[0088] In the data update submodule, new charge and discharge data is called based on the effect of the charge and discharge strategy to collect and preprocess data. The specific operation is to first use sensors and data acquisition systems to collect the latest charge and discharge data of low-speed electric vehicles in real time and store the data in a database. Then, noise filtering technology, such as Kalman filtering, is used to denoise the original data to remove random noise in the data. Statistical methods are used to detect and delete outliers, such as using the standard deviation method to mark data exceeding three times the standard deviation as abnormal and remove it. Next, the data is standardized and scaled to a range of 0 to 1 to ensure comparability between different data sources. Finally, the vehicle charge and discharge time point matrix is ​​updated, and the new data is added to the matrix to generate an updated charge and discharge matrix. The file generated by this submodule is an updated charge and discharge matrix file, which provides the latest charge and discharge data and provides a basis for subsequent similarity calculations and strategy optimization.

[0089] In the similarity recalculation submodule, the similarity between vehicles and the similarity between time points are recalculated based on the updated charge and discharge matrix. The specific operation is to first standardize the updated charge and discharge data, and use the Min-Max normalization method to scale the data to a uniform range. Then, the cosine similarity calculation method is used to calculate the similarity between each pair of vehicles. The specific steps are to represent the charge and discharge behavior data of each vehicle as a vector, calculate the cosine value between the vectors, and obtain the similarity matrix between vehicles. Next, the dynamic time warping (DTW) algorithm is used to calculate the similarity between different time points. The specific steps are to flexibly match the time series data, calculate its minimum distance, and generate a time point similarity matrix. Through the above content, a new similarity result file is generated, which contains the similarity matrix between vehicles and the time point similarity matrix, providing the latest similarity data support for strategy optimization.

[0090] In the strategy optimization submodule, the collaborative filtering model is adjusted based on the new similarity results to update the recommendation strategy. The specific operation is to first update the collaborative filtering model based on the new similarity results using matrix decomposition techniques such as singular value decomposition (SVD) and recalculate the potential feature matrix of users and items. Then, a genetic algorithm is used for strategy optimization. The specific steps are to perform crossover and mutation operations on the current recommendation strategy, generate multiple new strategy combinations, evaluate the applicability of each strategy, and select the optimal strategy for retention and update. Next, based on the optimized collaborative filtering model, the recommended charging and discharging strategy for each vehicle is recalculated to generate an optimized charging and discharging strategy file, which contains the adjusted charging time, discharging time, and updated charging frequency. Through the above operations, the continuous optimization of the recommended strategy is ensured, the battery efficiency is improved, and the battery life is extended. The effect of this submodule is to provide the latest and optimal charging and discharging strategy, providing an efficient solution for battery management of low-speed electric vehicles.

[0091] The energy storage power generation state evaluation method for a low-speed electric vehicle battery is performed based on the energy storage power generation state evaluation system for a low-speed electric vehicle battery, and includes the following steps:

[0092] S1: Based on the collected historical data of low-speed electric vehicle charging and discharging, a matrix of vehicle charging and discharging time points is established by using a matrix construction method. The similarity between vehicles is calculated to capture the vehicle group with the same charging and discharging behavior as the target vehicle. Then, the similarity between charging and discharging time points is calculated to capture the same charging and discharging time patterns and generate the same charging and discharging behavior.

[0093] S2: Based on similar charging and discharging behaviors, the cosine similarity analysis method is used to calculate the similarity between each vehicle, and the similarity of charging and discharging behaviors is analyzed through similarity to obtain a group of vehicles with similar charging and discharging behaviors as the target vehicle, and the similarity between multiple charging and discharging time points is calculated. The charging and discharging time pattern is analyzed through similarity to obtain the charging and discharging similarity analysis results;

[0094] S3: Based on the results of the charge and discharge similarity analysis, a scoring prediction method is used to combine the charge and discharge behaviors of similar vehicles to generate recommended charge and discharge time and frequency strategies. Multiple charge and discharge time points of the target vehicle are scored, and the optimal strategy is selected as the recommended strategy through scoring to generate the optimal charge and discharge strategy.

[0095] S4: Based on the optimal charging and discharging strategy, monitor the changes in vehicle battery performance and life, use data comparison and analysis methods, compare the battery status data before and after the application of the strategy, compare the charging amount, discharging amount and service life parameters, evaluate the strategy effect, obtain the charging and discharging strategy effect, call new charging and discharging data, update the vehicle charging and discharging time point matrix, recalculate the similarity between vehicles and between time points, adjust the collaborative filtering model, update the recommended strategy through the new similarity analysis results, and generate an optimized charging and discharging strategy.

[0096] This energy storage and power generation status assessment method for low-speed electric vehicle batteries has brought significant beneficial effects through systematic steps and precise algorithms. First, based on the collected historical data of low-speed electric vehicle charging and discharging, a vehicle charging and discharging time point matrix is ​​constructed, which can systematically capture and analyze the charging and discharging behavior data of a large number of vehicles. This matrix construction method helps to clearly display the charging and discharging patterns of different vehicles at different time points, providing a comprehensive charging and discharging behavior database, making subsequent similarity calculations and behavior pattern capture more accurate.

[0097] Secondly, by calculating the similarity between vehicles and the similarity between charging and discharging time points, it is possible to accurately identify groups of vehicles with similar charging and discharging behaviors to the target vehicle. By using the cosine similarity analysis method and conducting in-depth analysis of the group, it is possible to more accurately understand and predict charging and discharging behaviors. In this way, the system can extract valuable patterns from the charging and discharging behaviors of similar vehicles and generate more reasonable and efficient charging and discharging strategies for the target vehicle.

[0098] Furthermore, based on the similarity analysis results, the scoring prediction method and recommendation strategy are used to generate the optimal charging and discharging strategy, so that the charging and discharging operations of each vehicle are more optimized. By scoring multiple charging and discharging time points and selecting the optimal strategy, not only the charging efficiency is improved, but also the battery life is extended. This scoring prediction method ensures the scientificity and practicality of the recommendation strategy by combining big data and machine learning technology.

[0099] In addition, by monitoring and evaluating changes in vehicle battery performance and life, the actual effect of the charge and discharge strategy can be understood in real time. Using data comparison and analysis methods, the system can accurately judge the effect of the strategy, compare the battery status data before and after the strategy is applied, and ensure the continuous optimization of the strategy. By comparing the charge, discharge and service life parameters, the system can promptly discover and correct deficiencies in the charge and discharge strategy, continuously improve the charge and discharge operation, and improve overall efficiency.

[0100] Finally, new charging and discharging data is called, the vehicle charging and discharging time point matrix is ​​updated, and the similarities between vehicles and time points are recalculated to ensure the dynamic adjustment ability of the system. By adjusting the collaborative filtering model and updating the recommendation strategy, the system can continuously optimize the charging and discharging strategy to maintain the advancement and effectiveness of the strategy. In general, this method significantly improves the charging and discharging efficiency of low-speed electric vehicle batteries, extends the battery life, and realizes the intelligent and refined management of charging and discharging operations through systematic data collection and analysis combined with advanced algorithms and prediction methods.

[0101] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. An energy storage and power generation status assessment system for low-speed electric vehicle batteries, characterized in that: The system comprises: The charging and discharging behavior analysis module establishes a vehicle charging and discharging time point matrix based on the collected low-speed electric vehicle charging and discharging history data. The matrix elements represent the charging and discharging behavior of the target vehicle at the target time point. By calculating the similarity between the vehicles, the vehicle group with the same charging and discharging behavior as the target vehicle is captured. By calculating the similarity between the charging and discharging time points, the same charging and discharging time pattern is captured and the same charging and discharging behavior is generated. The collaborative filtering analysis module calculates the similarity between each vehicle based on the similar charging and discharging behaviors, analyzes the similarity of the charging and discharging behaviors by cosine similarity, obtains a group of vehicles with the same charging and discharging behaviors as the target vehicle, calculates the similarity between multiple charging and discharging time points, analyzes the charging and discharging time patterns by similarity, and obtains the charging and discharging similarity analysis results; An optimization strategy recommendation module generates a recommended charging and discharging time and frequency strategy based on the charging and discharging similarity analysis results and the charging and discharging behaviors of similar vehicles. It uses the scoring prediction technology to score multiple charging and discharging time points of the target vehicle, selects the strategy with the best score as the recommended strategy, and generates the optimal charging and discharging strategy. The model verification module applies the optimal charging and discharging strategy to the vehicle, monitors the performance and life of the vehicle battery, compares the battery status data before and after the strategy is applied, compares the charging amount, discharging amount and service life parameters, evaluates the strategy effect, and obtains the charging and discharging strategy effect; A strategy adjustment module, based on the effect of the charging and discharging strategy, calls new charging and discharging data, updates the vehicle charging and discharging time point matrix, recalculates the similarities between vehicles and between time points, adjusts the collaborative filtering model, updates the recommendation strategy through the new similarity analysis results, and generates an optimized charging and discharging strategy; The collaborative filtering analysis module includes: A similarity calculation submodule calculates the similarity between each vehicle based on the same type of charging and discharging behavior of each vehicle, analyzes the similarity of charging and discharging behavior through cosine similarity, uses the charging and discharging data of the vehicle to compare the charging start time and the end time, calculates the time difference between charging and discharging, and generates a vehicle similarity matrix; The charging and discharging behavior analysis submodule obtains a vehicle group with the same charging and discharging behavior as the target vehicle based on the vehicle similarity matrix, performs cluster analysis on the charging and discharging behavior of the vehicle group, and uses the charging frequency and duration of multiple vehicles in the group to perform clustering of similar features to obtain the characteristics of the vehicle group; The charging and discharging time pattern analysis submodule calculates the similarity between multiple charging and discharging time points based on the characteristics of the same type of vehicle group, analyzes the charging and discharging time pattern through similarity, uses time series comparison to identify the charging peak and valley time periods, and generates charging and discharging similarity analysis results.

2. The energy storage and power generation status assessment system for a low-speed electric vehicle battery according to claim 1, characterized in that: The similar charging and discharging behaviors include charging frequency similarity, discharge time similarity and charging duration similarity. The charging and discharging similarity analysis results include vehicle-to-vehicle similarity matrix, time point similarity matrix and similarity score list. The optimal charging and discharging strategy includes recommended charging time, recommended discharge time and recommended charging frequency. The charging and discharging strategy effects include battery charging amount change, battery discharge amount change and battery service life change. The optimized charging and discharging strategy includes adjusted charging time, adjusted discharge time and updated charging frequency.

3. The energy storage and power generation status assessment system for low-speed electric vehicle batteries according to claim 1, characterized in that: The charge and discharge behavior analysis module includes: The data collection submodule performs data cleaning and preprocessing based on the collected historical data of low-speed electric vehicle charging and discharging, removes noise and outliers, and establishes a vehicle charging and discharging time point matrix. The matrix elements represent the charging and discharging behavior of the target vehicle at the target time point, and generates a charging and discharging time point matrix; A similarity calculation submodule performs data standardization processing based on the charge and discharge time point matrix, calculates the similarity of the charge and discharge behaviors between vehicles, identifies vehicles with similar charge and discharge behaviors, captures a group of vehicles with similar charge and discharge behaviors as the target vehicle, and generates a group of similar vehicles; The behavior pattern generation submodule performs cluster analysis on the charging and discharging time point data based on the same type of vehicle group, identifies the charging and discharging time pattern, captures the same type of charging and discharging time pattern, and generates the same type of charging and discharging behavior.

4. The energy storage and power generation status assessment system for low-speed electric vehicle batteries according to claim 1, characterized in that: The optimization strategy recommendation module includes: A strategy generation submodule, based on the charge and discharge similarity analysis results and combined with the charge and discharge behaviors of similar vehicles, collects and analyzes data to perform statistics and optimization on the charge and discharge time and frequency, and generates a recommended charge and discharge strategy; The scoring prediction submodule scores multiple charging and discharging time points of the target vehicle based on the recommended charging and discharging strategy, calculates the scoring value of each time point through data analysis and prediction, and generates a charging and discharging scoring result; The strategy selection submodule selects the best-scoring charge and discharge strategy based on the charge and discharge scoring results, determines the best solution through comparison and selection, and uses it as the final recommended strategy to generate the best charge and discharge strategy.

5. The energy storage and power generation status assessment system for low-speed electric vehicle batteries according to claim 1, characterized in that: The model verification module includes: A strategy application submodule, based on the optimal charging and discharging strategy, applies the strategy to the vehicle, monitors the vehicle status during the strategy execution, records the time points of charging and discharging, uses an on-board data recorder to record the charging voltage and current, monitors the start and end time of each charging and discharging, and generates a strategy application record; The battery monitoring submodule monitors the performance and life of the vehicle battery based on the strategy application record, uses sensors to collect the voltage, current and temperature data of the battery, and records the change in battery capacity during each charge and discharge, regularly checks the internal resistance and voltage stability of the battery, and generates battery performance data; The effect evaluation submodule compares the battery status before and after the application of the strategy based on the battery performance data, calculates the charging capacity, discharging capacity and service life parameters, uses the data comparison method to judge the strategy effect, compares the battery charging and discharging efficiency and life changes in multiple time periods, and obtains the charging and discharging strategy effect.

6. The energy storage and power generation status assessment system for low-speed electric vehicle batteries according to claim 1, characterized in that: The policy adjustment module includes: The data update submodule calls new charge and discharge data based on the effect of the charge and discharge strategy, performs data collection and preprocessing, removes noise and outliers, updates the vehicle charge and discharge time point matrix, and generates an updated charge and discharge matrix; A similarity recalculation submodule, based on the updated charge-discharge matrix, recalculates the similarity between vehicles and the similarity between time points, and generates a new similarity result through data standardization processing and similarity calculation; The strategy optimization submodule adjusts the collaborative filtering model based on the new similarity result, updates the recommendation strategy, and generates an optimized charging and discharging strategy through strategy optimization and adjustment.

7. A method for evaluating the energy storage and power generation status of a low-speed electric vehicle battery, characterized in that: The energy storage and power generation status assessment system for a low-speed electric vehicle battery according to any one of claims 1 to 6 comprises the following steps: Based on the collected historical data of low-speed electric vehicle charging and discharging, a matrix of vehicle charging and discharging time points is established by using a matrix construction method. The similarity between vehicles is calculated to capture the vehicle group with the same charging and discharging behavior as the target vehicle. Then, the similarity between charging and discharging time points is calculated to capture the same charging and discharging time patterns and generate the same charging and discharging behavior. Based on the similar charging and discharging behaviors, a cosine similarity analysis method is used to calculate the similarity between each vehicle, and the similarity of the charging and discharging behaviors is analyzed by similarity to obtain a group of vehicles with the same charging and discharging behaviors as the target vehicle, and the similarity between multiple charging and discharging time points is calculated. The charging and discharging time pattern is analyzed by similarity to obtain a charging and discharging similarity analysis result; Based on the charge and discharge similarity analysis results, a scoring prediction method is used to combine the charge and discharge behaviors of similar vehicles to generate recommended charge and discharge time and frequency strategies, score multiple charge and discharge time points of the target vehicle, select the best strategy as the recommended strategy through scoring, and generate the best charge and discharge strategy; Based on the optimal charging and discharging strategy, the vehicle battery performance and life changes are monitored, and a data comparison and analysis method is used to compare the battery status data before and after the application of the strategy, and compare the charging amount, discharging amount and service life parameters to evaluate the strategy effect, obtain the charging and discharging strategy effect, call new charging and discharging data, update the vehicle charging and discharging time point matrix, recalculate the similarity between vehicles and between time points, adjust the collaborative filtering model, update the recommendation strategy through the new similarity analysis results, and generate an optimized charging and discharging strategy.

Images