Dynamic analysis charging pile planning method

Abstract

The invention relates to the technical field of power grid planning, in particular to a charging pile planning method based on dynamic analysis. The method comprises the following steps: obtaining electric vehicle use data in a target area, and carrying out charging demand prediction analysis according to the electric vehicle use data to obtain a charging demand prediction model; based on the charging demand prediction model, performing multi-scene simulation to obtain charging demand distribution map data, multiple scenes including different traffic flows, different driving modes and different charging demand scenes; obtaining geographic information data of the target area; and constructing a spatial analysis model according to the geographic information data, and evaluating the candidate position of the charging pile according to the charging demand distribution map data, thereby obtaining position suitability score data. According to the method, different scenes are simulated, so that the adaptability and robustness of planning are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid planning, and in particular to a charging pile planning method based on dynamic analysis. Background Art

[0002] In recent years, with the rapid popularization of electric vehicles (EVs), the planning and construction of charging infrastructure has become a key issue in urban development and power grid management. In the charging pile planning method, dynamic analysis specifically refers to the continuous and real-time analysis and optimization of multiple data such as charging demand, geographical location, and power grid load at different times and scenarios to ensure that the planning scheme can adapt to the dynamic actual needs and conditions. This includes the following aspects: Dynamic demand forecasting: Based on real-time or historical electric vehicle usage data, predict the charging demand in different time periods (such as days, weeks, seasons, etc.) and different scenarios (such as peak periods, trough periods, etc.); this forecast needs to consider multiple factors such as traffic flow, driving mode, charging habits, and dynamically adjust according to actual conditions. Dynamic feedback and adjustment: Establish a multi-level feedback control system to dynamically adjust and optimize parameters through real-time monitoring and feedback of charging pile usage and power grid load to adapt to changing needs and environments.

[0003] Traditional charging pile planning methods often rely on static data and single-scenario analysis, which makes it difficult to adapt to the rapidly changing electric vehicle usage environment and complex transportation networks. This method has some obvious limitations when facing dynamic traffic flow, uncertain driving patterns, and changing charging needs, leading to problems such as low charging pile utilization, poor user experience, and unbalanced grid load. Summary of the invention

[0004] Based on this, it is necessary for the present invention to provide a dynamic analysis charging pile planning method to solve at least one of the above technical problems.

[0005] To achieve the above purpose, a dynamic analysis charging pile planning method includes the following steps: Step S1: acquiring electric vehicle usage data in the target area, performing charging demand forecasting analysis based on the electric vehicle usage data, and obtaining a charging demand forecasting model; performing multi-scenario simulation based on the charging demand forecasting model, and obtaining charging demand distribution map data, wherein the multi-scenario includes scenarios with different traffic flows, different driving modes, and different charging demands; Step S2: Obtain geographic information data of the target area; construct a spatial analysis model based on the geographic information data, and evaluate candidate locations of charging piles based on the charging demand distribution map data, thereby obtaining location suitability score data; Step S3: Establish a charging pile performance-cost model based on the pre-acquired technical parameters and cost data of different types of charging piles, and perform a multi-objective optimization analysis based on cost-coverage-service quality based on the location suitability score data to obtain a preliminary configuration plan; Step S4: constructing a power system capacity model according to the power grid information of the target area; analyzing the power grid carrying capacity of the preliminary configuration plan according to the power system capacity model, and adjusting the configuration plan to obtain a feasible configuration plan; dynamically simulating the feasible configuration plan, and performing a performance evaluation based on the utilization rate, waiting time and power grid load to obtain performance evaluation data; Step S5: According to the preset performance target and performance evaluation data, the feasible configuration scheme is dynamically optimized and adjusted driven by the performance target, so as to obtain the final dynamic planning scheme.

[0006] By collecting and analyzing the usage data of electric vehicles, the present invention can accurately predict the charging demand in different time periods and different scenarios, and ensure the scientificity and reliability of the planning scheme. Through multi-scenario simulation, taking into account the complex situations such as different traffic flows, driving modes and charging needs, it is ensured that the charging pile layout scheme can adapt to the changing actual situation and provide efficient services. By using geographic information system (GIS) technology, combined with charging demand distribution data, the candidate locations are scientifically evaluated to ensure the rationality of charging pile layout and improve the utilization rate and convenience. By scoring the location suitability, the best layout location is selected to maximize the service range, improve user experience and charging pile utilization efficiency. By establishing a performance-cost model, the technical parameters and costs of different types of charging piles are comprehensively considered to ensure that the construction cost is controlled while meeting the performance requirements; a multi-objective optimization analysis of cost, coverage and service quality is carried out to ensure that the preliminary configuration scheme achieves the optimal balance between cost, service quality and coverage. Through the power system capacity model, the carrying capacity of the power grid for the charging pile configuration scheme is analyzed to avoid power grid overload and ensure the stability and safety of the power system; through dynamic simulation, performance indicators such as utilization rate, waiting time and power grid load under different load conditions are evaluated to ensure the feasibility and efficiency of the scheme in practical applications. According to the performance evaluation data and preset goals, the feasible solutions are dynamically optimized and adjusted to ensure that the final solution can meet the performance goals to the maximum extent, improve the utilization efficiency of the charging piles and user satisfaction; the dynamic optimization process enables the solution to be continuously adjusted and optimized according to real-time data and actual conditions, ensuring that the charging pile network is always in the best operating state. Through the effects of the above steps, the charging pile planning method of the present invention can achieve a scientific and reasonable layout of charging piles, meet the ever-changing actual needs, improve the utilization efficiency of charging piles, and ensure the safety of the power grid and user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings: Figure 1 A schematic diagram of the steps of the charging pile planning method for dynamic analysis of the present invention; Figure 2 for Figure 1 Detailed step flow diagram of step S1; Figure 3 for Figure 1 Detailed step flow chart of step S2 in FIG. DETAILED DESCRIPTION

[0008] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0009] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0010] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0011] To achieve this, please refer to Figures 1 to 3 The present invention provides a charging pile planning method for dynamic analysis, the method comprising the following steps: Step S1: acquiring electric vehicle usage data in the target area, performing charging demand forecasting analysis based on the electric vehicle usage data, and obtaining a charging demand forecasting model; performing multi-scenario simulation based on the charging demand forecasting model, and obtaining charging demand distribution map data, wherein the multi-scenario includes scenarios with different traffic flows, different driving modes, and different charging demands; Step S2: Obtain geographic information data of the target area; construct a spatial analysis model based on the geographic information data, and evaluate candidate locations of charging piles based on the charging demand distribution map data, thereby obtaining location suitability score data; Step S3: Establish a charging pile performance-cost model based on the pre-acquired technical parameters and cost data of different types of charging piles, and perform a multi-objective optimization analysis based on cost-coverage-service quality based on the location suitability score data to obtain a preliminary configuration plan; Step S4: constructing a power system capacity model according to the power grid information of the target area; analyzing the power grid carrying capacity of the preliminary configuration plan according to the power system capacity model, and adjusting the configuration plan to obtain a feasible configuration plan; dynamically simulating the feasible configuration plan, and performing a performance evaluation based on the utilization rate, waiting time and power grid load to obtain performance evaluation data; Step S5: According to the preset performance target and performance evaluation data, the feasible configuration scheme is dynamically optimized and adjusted driven by the performance target, so as to obtain the final dynamic planning scheme.

[0012] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of a dynamic analysis charging pile planning method of the present invention. In this example, the dynamic analysis charging pile planning method includes the following steps: Step S1: acquiring electric vehicle usage data in the target area, performing charging demand forecasting analysis based on the electric vehicle usage data, and obtaining a charging demand forecasting model; performing multi-scenario simulation based on the charging demand forecasting model, and obtaining charging demand distribution map data, wherein the multi-scenario includes scenarios with different traffic flows, different driving modes, and different charging demands; The embodiment of the present invention collects the usage data of electric vehicles in the target area, including vehicle type, driving mode, charging habits, etc. The data source may include self-reported data from car owners, vehicle system data, traffic monitoring system data, etc.; the collected data is preprocessed to remove outliers and noise, fill in missing values, and ensure data quality. Use statistical methods or machine learning algorithms (such as time series analysis, regression analysis, neural networks, etc.) to establish a charging demand prediction model. Set different scenario parameters (such as traffic flow, driving mode, charging demand type) for simulation and generate a charging demand distribution map, which may include demand forecasts under different scenarios such as morning peak and evening peak, short-distance and long-distance driving in cities; according to the simulation results, draw a charging demand distribution map to identify high-demand areas and low-demand areas.

[0013] Step S2: Obtain geographic information data of the target area; construct a spatial analysis model based on the geographic information data, and evaluate candidate locations of charging piles based on the charging demand distribution map data, thereby obtaining location suitability score data; The embodiment of the present invention collects geographic information data of the target area, including terrain, road network, land use, population density, existing charging facilities and power grid distribution, etc., processes errors, missing values ​​and duplicate records in the data, and integrates data from different sources into a unified geographic information system (GIS) platform. Analyze the impact of terrain characteristics on the layout of charging piles, such as slope, altitude, etc., evaluate the impact of road network on the accessibility of charging piles, including shortest path analysis, traffic flow analysis, etc., determine the suitability of land use for the layout of charging piles, such as commercial areas, residential areas, industrial areas, etc., analyze the impact of population density distribution on charging demand, and based on the above analysis, evaluate candidate locations for charging piles and generate location suitability score data.

[0014] Step S3: Establish a charging pile performance-cost model based on the pre-acquired technical parameters and cost data of different types of charging piles, and perform a multi-objective optimization analysis based on cost-coverage-service quality based on the location suitability score data to obtain a preliminary configuration plan; The embodiment of the present invention obtains technical parameters (such as charging power, efficiency, interface type) and cost data (including equipment cost, installation cost, operation cost, and maintenance cost) of different types of charging piles. Based on the technical parameters and cost data, a charging pile performance-cost model is established, and the accuracy of the model is verified using historical data or simulation data. Optimization goals are set, including minimizing total cost, maximizing coverage, maximizing service quality, etc.; optimization algorithms (such as genetic algorithms, particle swarm optimization, etc.) are used to analyze location suitability score data to obtain a preliminary configuration plan.

[0015] Step S4: constructing a power system capacity model according to the power grid information of the target area; analyzing the power grid carrying capacity of the preliminary configuration plan according to the power system capacity model, and adjusting the configuration plan to obtain a feasible configuration plan; dynamically simulating the feasible configuration plan, and performing a performance evaluation based on the utilization rate, waiting time and power grid load to obtain performance evaluation data; The embodiment of the present invention collects the power grid information of the target area, including the distribution of substations, the layout of power lines, the load distribution, etc., establishes a topological structure model of the power grid, describes the node and line information, establishes a capacity limitation model of the line and substation, considers factors such as thermal capacity and stability, integrates the topological structure model and the capacity limitation model, and constructs a complete power system capacity model. Use the power system simulation tool to analyze the power grid load on the preliminary configuration plan, and adjust the charging pile configuration that does not meet the power grid carrying capacity according to the analysis results to ensure that the power grid load is within a safe range. Design a simulation of the operating status of the charging station in different time periods and scenarios, and perform performance evaluation based on utilization rate, waiting time and power grid load to obtain performance evaluation data.

[0016] Step S5: According to the preset performance target and performance evaluation data, the feasible configuration scheme is dynamically optimized and adjusted driven by the performance target, so as to obtain the final dynamic planning scheme.

[0017] The embodiment of the present invention defines performance evaluation criteria according to preset performance targets, such as the utilization rate of charging piles, waiting time, grid load, etc., compares the performance evaluation data with the preset performance targets, and identifies the gap. With the performance target difference data as the constraint condition and the minimization of adjustment cost as the objective function, a dynamic optimization model is established; an optimization algorithm is designed to consider the adjustment of the number of charging piles, the replacement of the type, and the fine-tuning of the location. The optimization algorithm is applied to optimize and adjust the feasible configuration scheme to obtain the final dynamic planning scheme to meet the performance targets and optimization requirements.

[0018] By collecting and analyzing the usage data of electric vehicles, the present invention can accurately predict the charging demand in different time periods and different scenarios, and ensure the scientificity and reliability of the planning scheme. Through multi-scenario simulation, taking into account the complex situations such as different traffic flows, driving modes and charging needs, it is ensured that the charging pile layout scheme can adapt to the changing actual situation and provide efficient services. By using geographic information system (GIS) technology, combined with charging demand distribution data, the candidate locations are scientifically evaluated to ensure the rationality of charging pile layout and improve the utilization rate and convenience. By scoring the location suitability, the best layout location is selected to maximize the service range, improve user experience and charging pile utilization efficiency. By establishing a performance-cost model, the technical parameters and costs of different types of charging piles are comprehensively considered to ensure that the construction cost is controlled while meeting the performance requirements; a multi-objective optimization analysis of cost, coverage and service quality is carried out to ensure that the preliminary configuration scheme achieves the optimal balance between cost, service quality and coverage. Through the power system capacity model, the carrying capacity of the power grid for the charging pile configuration scheme is analyzed to avoid power grid overload and ensure the stability and safety of the power system; through dynamic simulation, performance indicators such as utilization rate, waiting time and power grid load under different load conditions are evaluated to ensure the feasibility and efficiency of the scheme in practical applications. According to the performance evaluation data and preset goals, the feasible solutions are dynamically optimized and adjusted to ensure that the final solution can meet the performance goals to the maximum extent, improve the utilization efficiency of the charging piles and user satisfaction; the dynamic optimization process enables the solution to be continuously adjusted and optimized according to real-time data and actual conditions, ensuring that the charging pile network is always in the best operating state. Through the effects of the above steps, the charging pile planning method of the present invention can achieve a scientific and reasonable layout of charging piles, meet the ever-changing actual needs, improve the utilization efficiency of charging piles, and ensure the safety of the power grid and user satisfaction.

[0019] Preferably, step S11 includes the following steps: Step S11: Acquire electric vehicle usage data in the target area, including vehicle type data, driving mode data, and charging demand data; Step S12: standardizing the electric vehicle usage data and processing outliers to obtain electric vehicle standard data; Step S13: extracting driving mode features based on average driving distance and frequency from the electric vehicle standard data, thereby obtaining driving mode feature data; Step S14: extracting charging behavior characteristics based on charging frequency, charging amount, charging type and charging time from the electric vehicle standard data, thereby obtaining charging behavior characteristic data; Step S15: clustering the driving mode characteristic data and the charging behavior characteristic data into different types of vehicle characteristics according to the vehicle type data, thereby obtaining vehicle type usage characteristic data; Step S16: performing charging demand forecasting analysis based on the driving mode characteristic data, the charging behavior characteristic data, and the vehicle type usage characteristic data to obtain a charging demand forecasting model; Step S17: Perform multi-scenario simulation based on the charging demand prediction model to obtain charging demand distribution map data, wherein the multi-scenario includes scenarios with different traffic flows, different driving modes, and different charging demands.

[0020] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1 Detailed step flow diagram of step S1 in the embodiment of the present invention, step S1 includes the following steps: Step S11: Acquire electric vehicle usage data in the target area, including vehicle type data, driving mode data, and charging demand data; The embodiments of the present invention collect basic information about electric vehicles, such as vehicle model, age, battery capacity, etc. These data can usually be obtained from vehicle registration records, owner reports or vehicle systems. Obtain vehicle driving mode data, including driving frequency, driving time, driving distance, etc., which can be obtained through on-board GPS, driving recorders and other devices. Record the charging time, charging amount, charging location, etc. of each electric vehicle. The data sources include charging pile records, user manual input, etc. Integrate data from different sources into a unified data management platform to ensure data integrity and consistency.

[0021] Step S12: standardizing the electric vehicle usage data and processing outliers to obtain electric vehicle standard data; The embodiment of the present invention unifies the metric standards of different data sources, such as converting distance into kilometers and standardizing time into hours; standardizes the data (such as Z-score standardization) to make them have the same scale for comparison and analysis. Outliers are identified through statistical analysis (such as box plots, Z-scores), and outliers are processed, such as excluding, adjusting or replacing, to ensure the accuracy and reliability of the data set.

[0022] Step S13: extracting driving mode features based on average driving distance and frequency from the electric vehicle standard data, thereby obtaining driving mode feature data; The embodiment of the present invention calculates the average driving distance of each vehicle (for example, the average daily driving distance, the average monthly driving distance), and calculates the driving frequency of each vehicle, such as the number of driving days per week or the number of driving times per month; takes the average driving distance and driving frequency as main features to form driving pattern feature data, and organizes the feature data into a data format suitable for subsequent analysis.

[0023] Step S14: extracting charging behavior characteristics based on charging frequency, charging amount, charging type and charging time from the electric vehicle standard data, thereby obtaining charging behavior characteristic data; The embodiment of the present invention calculates the charging frequency of each vehicle (for example, the number of charging times per month), records the charging amount of each charging, and calculates the total charging amount or the average charging amount; records the charging type, such as fast charging, slow charging, etc., and counts the usage of different types of charging, records the duration of each charging, and calculates the total charging time or the average charging time. The charging frequency, charging amount, charging type and charging time are organized into charging behavior feature data for subsequent analysis.

[0024] Step S15: clustering the driving mode characteristic data and the charging behavior characteristic data into different types of vehicle characteristics according to the vehicle type data, thereby obtaining vehicle type usage characteristic data; The embodiment of the present invention integrates vehicle type data with driving mode characteristic data and charging behavior characteristic data, selects a suitable clustering algorithm (such as K-means, hierarchical clustering, DBSCAN, etc.), performs clustering according to vehicle type, driving mode characteristics and charging behavior characteristics, identifies different types of vehicle characteristic groups, analyzes the usage characteristics of different types of vehicles, and determines the typical behavior patterns of various types of vehicles.

[0025] Step S16: performing charging demand forecasting analysis based on the driving mode characteristic data, the charging behavior characteristic data, and the vehicle type usage characteristic data to obtain a charging demand forecasting model; The embodiment of the present invention prepares driving mode characteristic data, charging behavior characteristic data and vehicle type use characteristic data as input features of the prediction model; selects a suitable prediction model (such as linear regression, support vector machine, decision tree, neural network, etc.); uses historical data to train the prediction model and adjusts the model parameters to improve the prediction accuracy. The trained prediction model is used to predict future charging demand, and the prediction results are analyzed to determine the charging demand under different circumstances.

[0026] Step S17: Perform multi-scenario simulation based on the charging demand prediction model to obtain charging demand distribution map data, wherein the multi-scenario includes scenarios with different traffic flows, different driving modes, and different charging demands.

[0027] The embodiment of the present invention sets different scenario parameters, including traffic flow (morning peak, evening peak, uniform throughout the day), driving mode (urban short distance, urban and suburban mixed, long-distance high-speed), and charging demand (family night, public charging, mixed demand). Based on the charging demand prediction model, the charging demand in different scenarios is simulated, and the charging demand prediction results in each scenario are recorded; the simulation results are plotted into a charging demand distribution map to identify charging demand hotspots and low demand areas in different scenarios.

[0028] The present invention obtains comprehensive electric vehicle usage data, covering vehicle types, driving modes and charging requirements, provides basic data support for subsequent analysis, and ensures the accuracy and reliability of the prediction results. Through standardization and outlier processing, the noise and deviation in the data are eliminated, the consistency and accuracy of the data are ensured, and the credibility of the subsequent analysis results is improved. By extracting driving mode characteristics such as average driving distance and frequency, the driving habits and modes of electric vehicles can be accurately identified, providing an important basis for charging demand prediction. By analyzing charging behavior characteristics such as charging frequency, charging amount, charging type and charging time, the charging habits of electric vehicle users can be deeply understood, and a charging pile planning scheme that is closer to actual needs can be formulated. By clustering analysis of driving modes and charging behaviors of different vehicle types, the usage characteristics of different types of vehicles can be refined, and targeted solutions can be provided for the charging needs of different types of vehicles. Combining driving modes, charging behaviors and vehicle type characteristics, a charging demand prediction model is constructed, which can more accurately predict future charging needs and provide a scientific basis for decision-making. Through multi-scenario simulation, the distribution of charging demand under different traffic flows, driving modes and charging demands can be considered to ensure that the planning scheme can operate efficiently under various circumstances; the charging demand distribution map data provides a basis for the reasonable layout of charging piles, ensuring that the distribution of charging piles can meet actual needs to the greatest extent, improve user satisfaction and charging pile utilization efficiency. Through the above steps, this charging pile planning method can systematically acquire, process and analyze the usage data of electric vehicles, accurately predict future charging needs, and perform multi-scenario simulation. This method not only improves the accuracy of data processing and the comprehensiveness of analysis, but also provides a scientific basis for the reasonable layout and optimization of charging piles, ensuring that the planning scheme can adapt to the changing actual situation, meet user needs to the greatest extent, and improve the utilization efficiency of charging piles and user satisfaction.

[0029] Preferably, step S16 comprises the following steps: Step S161: merging the driving mode characteristic data, the charging behavior characteristic data and the vehicle type usage characteristic data into a data set, and performing a data integrity check and missing value processing to obtain a comprehensive characteristic data set; The embodiment of the present invention combines the driving mode characteristic data, charging behavior characteristic data and vehicle type usage characteristic data into a comprehensive characteristic data set according to the vehicle ID or other unique identifier, ensuring that the format of the merged data is consistent, such as column name, data type, etc. Use statistical methods (such as the isnull() method) to check missing values ​​in the data set, ensure that the fields of all records conform to the expected format and range, and check whether there are abnormal values ​​or inconsistent records. Use appropriate methods (such as mean filling, median filling, forward filling or interpolation) to fill missing values. If there are too many missing values ​​or the filling is inappropriate, records with more missing values ​​can be deleted.

[0030] Step S162: performing time series processing on the comprehensive feature data set, and performing trend, seasonality and random component decomposition, thereby obtaining comprehensive feature time series data; The embodiment of the present invention sorts the comprehensive feature data set in chronological order to ensure the correctness of the chronological order; defines a time window (such as hour, day, week) to ensure that the time granularity of the data is suitable for analysis. Use a time series decomposition method (such as STL decomposition, Holt-Winters method) to decompose the time series data into trend, seasonal and random components; select a suitable model to decompose the time series data to obtain trend components, seasonal components and random components. The decomposition results of trend, seasonal and random components are aggregated into comprehensive feature time series data for further analysis.

[0031] Step S163: Calculate the correlation coefficient of the comprehensive characteristic time series data and the charging demand data to obtain correlation coefficient data, wherein the correlation coefficient calculation includes calculating the linear correlation between continuous variables by using the Pearson correlation coefficient, calculating the nonlinear correlation of categorical variables by using the Spearman rank correlation coefficient, and calculating the time lag correlation by using the cross correlation function; The embodiment of the present invention uses the Pearson correlation coefficient (PearsonR) to calculate the linear correlation between the comprehensive feature time series data and the charging demand data; analyzes the Pearson correlation coefficient value (between -1 and 1) to understand the linear relationship between the feature and the demand. Use the Spearman rank correlation coefficient (SpearmanR) to calculate the nonlinear correlation between the categorical variable or ranking data and the charging demand data; analyzes the Spearman coefficient value (between -1 and 1) to understand the relationship between the categorical variable or ranking data and the demand. Use the cross-correlation function (CCF) to calculate the lagged correlation of the time series data, analyze the correlation under different time lags, and understand the relationship between the time series data.

[0032] Step S164: performing feature selection based on Gini impurity on the comprehensive feature data set according to the correlation coefficient data, thereby obtaining a demand key feature set; The Gini impurity in the embodiment of the present invention is used to evaluate the splitting quality of features. The lower the Gini coefficient, the higher the splitting quality of the feature. The Gini impurity of each feature is calculated to identify the most useful features for charging demand prediction. The features are sorted according to the Gini impurity, and the features with the lowest Gini impurity are selected as the demand key feature set. The features are screened according to the selection criteria to construct the demand key feature set for subsequent model construction.

[0033] Step S165: Using the demand key feature set as input features and the charging demand data as the target variable, constructing and training a decision tree model, and optimizing the model parameters through cross-validation to obtain a charging demand prediction model.

[0034] The embodiment of the present invention selects a decision tree model as a prediction tool, uses a set of key demand features as input features, and uses charging demand data as the target variable for model training. The model is verified using k-fold cross validation (such as k=5 or k=10), and the model performance is optimized by adjusting the hyperparameters of the decision tree (such as the depth of the tree and the minimum number of sample leaves); the prediction performance of the model is evaluated using indicators such as accuracy, mean square error (MSE), and mean absolute error (MAE), and the model parameters are adjusted or feature engineering improvements are performed based on the evaluation results to improve the prediction accuracy.

[0035] The present invention merges data sets from different sources, performs data integrity checks and missing value processing, ensures the integrity and consistency of the data set, and improves the accuracy and reliability of subsequent analysis; the processed comprehensive feature data set reduces data noise and incompleteness, making subsequent modeling and analysis more accurate and effective. Through time series processing, trends, seasonality and random components in the data can be identified, providing a deep understanding of data changes, helping the prediction model to better capture the dynamic changes in charging demand; the decomposed time series data can more clearly show the structure of the data, helping the model to better understand and predict future charging demand. Through a variety of correlation coefficient calculation methods, the relationship between features and charging demand is analyzed from different angles, including linear correlation, nonlinear correlation and time lag correlation, to fully understand the impact of features on charging demand; correlation coefficient data helps identify key factors affecting charging demand, providing a basis for subsequent feature selection and model construction. Through feature selection, redundant features and irrelevant features are reduced, the training speed and efficiency of the model are improved, and the risk of overfitting is reduced; the selected key feature set can better explain the changes in charging demand and enhance the explanatory power and credibility of the model. The decision tree model can capture the complex relationship between features and charging demand, and optimize the model parameters through cross-validation to improve the performance and accuracy of the prediction model; by selecting key feature sets and optimizing model parameters, the prediction model can more accurately predict future charging demand and provide a more scientific and effective planning basis. Through the above steps, the driving mode, charging behavior and vehicle type feature data of electric vehicles can be systematically integrated, analyzed and processed, and finally an efficient charging demand prediction model can be constructed. These steps not only improve the quality of data processing and the comprehensiveness of analysis, but also ensure that the model can accurately capture the dynamic changes of charging demand through multi-angle correlation analysis and feature selection, and provide scientific and reliable decision support for charging pile planning.

[0036] Preferably, step S17 comprises the following steps: Step S171: Design multiple scenarios based on different traffic flows, different driving modes, and different charging demands according to the charging demand prediction model, and parameterize the scenarios to obtain scenario parameter setting data, wherein different traffic flows include increased charging demand during the morning peak, increased charging demand during the evening peak, and uniform charging demand throughout the day; different driving modes include mainly short-distance driving within the city, mixed driving in the city and suburbs, and mainly long-distance high-speed driving; different charging demands include mainly charging at home at night, mainly charging at public charging piles, and mixed charging demands; The embodiment of the present invention designs different situations of traffic flow, including morning peak (significant increase in traffic flow), evening peak (significant increase in traffic flow) and traffic evenly distributed throughout the day; defines driving modes, including short-distance driving in the city, mixed driving in the city and suburbs, and long-distance high-speed driving; sets different scenarios of charging demand, including mainly charging at home at night, mainly charging at public charging piles, and mixed charging demand modes. Define a specific parameter range or value for each scenario variable. For example, the morning peak flow can be set to increase the flow by 30% in a certain time period; generate different scenario configurations for each combination of traffic flow, driving mode and charging demand, and each configuration will correspond to a specific parameter setting data set for subsequent simulation and analysis. Integrate the parameters of all scenario variables into scenario parameter setting data, including charging demand, traffic mode, driving mode, etc. under different scenarios, to ensure that the scenario parameter setting data is structured for subsequent simulation and analysis.

[0037] Step S172: performing Monte Carlo simulation on the scenario parameter setting data to generate charging scenario simulation data; The embodiment of the present invention determines the number of Monte Carlo simulation runs (e.g., 1000 times) to ensure the reliability and statistical significance of the results, and defines the distribution of random variables (e.g., normal distribution, uniform distribution) based on scenario parameter setting data to simulate changes in charging demand. A random number generator is used to generate a large number of random samples according to the defined distribution to simulate charging demand under different scenarios, and charging demand prediction calculations are performed on each random sample to generate charging scenario simulation data. The results of the Monte Carlo simulation are summarized to generate charging scenario simulation data sets. These data sets will contain charging demand prediction results under different scenarios.

[0038] Step S173: inputting the charging scenario simulation data into the charging demand prediction model to obtain charging demand prediction result data under different scenarios; The embodiment of the present invention loads the generated charging scenario simulation data into the charging demand prediction model to ensure that the format of the simulation data meets the input requirements of the prediction model. The charging demand prediction model is used to predict the charging scenario simulation data, and the prediction result data output by the model is recorded and sorted to obtain the charging demand prediction result data under different scenarios. The prediction result data under different scenarios are aggregated to generate a complete data set of the charging demand prediction results, ensuring the integrity and accuracy of the prediction result data, and providing a reliable data basis for subsequent analysis.

[0039] Step S174: Perform a charging demand change analysis based on the time dimension on the charging demand prediction result data, perform a spatial clustering analysis based on the charging demand quantity, and identify charging demand hot spots, thereby obtaining charging demand distribution map data.

[0040] The embodiment of the present invention organizes the charging demand forecast result data by time dimension, such as by hour, day or week, analyzes the change trend, periodicity and outliers of the time series data, and identifies the time change pattern of the charging demand; uses spatial clustering algorithms (such as K-means, DBSCAN, hot spot analysis) to perform spatial clustering analysis on the charging demand forecast result data; clusters the charging demand data according to geographical location to identify areas with higher demand. The hot spots of charging demand are identified through the cluster analysis results, which represent geographical locations with relatively high charging demand; the hot spots of charging demand and demand distribution data are visualized to generate charging demand distribution map data for decision support.

[0041] The present invention ensures that the charging pile planning can adapt to different actual usage situations and improve the applicability and flexibility of the solution by designing a variety of scenarios of traffic flow, driving mode and charging demand; parameterizing the scenarios can refine the scenario design, ensure the accuracy and scientificity of the simulation, and provide a solid foundation for subsequent simulation and analysis. Monte Carlo simulation can effectively deal with uncertainty, and obtain more reliable and comprehensive simulation data by simulating charging demand under different scenarios through a large number of random sampling; generating a large amount of charging scenario simulation data, enriching the data basis of the charging demand prediction model, and improving the reliability and accuracy of the prediction results. Inputting the simulation data into the charging demand prediction model can make full use of the results of the scenario simulation and improve the accuracy and credibility of the charging demand prediction; obtaining the charging demand prediction results under different scenarios, it is convenient to compare and analyze different scenarios and find the best planning scheme. By analyzing the changes in the prediction results in the time dimension, we can identify the changing rules of charging demand in different periods and provide a basis for the dynamic management and optimization of charging piles. Based on the spatial clustering analysis of charging demand, we can identify the hot spots of charging demand, which is helpful to optimize the layout of charging piles and improve the utilization rate and service quality of charging facilities. We can generate charging demand distribution map data and visualize the spatial distribution of charging demand to facilitate intuitive understanding and decision-making. Through the above steps, we can fully consider the scenarios of different traffic flows, driving modes and charging demands, deal with uncertainties through Monte Carlo simulation, generate rich scenario simulation data, and use it in the charging demand prediction model to obtain accurate prediction results. By analyzing the prediction results in the time and space dimensions, we can identify the changing rules and hot spots of charging demand and provide a scientific basis for the layout of charging piles. These steps not only improve the accuracy and reliability of charging demand prediction, but also provide comprehensive and scientific support for the dynamic planning and optimization of charging piles through data visualization and scenario analysis.

[0042] Preferably, step S2 comprises the following steps: Step S21: Acquire geographic information data of the target area, including terrain data, road network data, land use data, population density data, existing charging facility distribution data, and power grid distribution data; Step S22: preprocessing and integrating the geographic information data to obtain a comprehensive geographic information data set, wherein the data preprocessing includes data cleaning, format unification, and coordinate system conversion; Step S23: constructing a spatial analysis model based on the geographic information comprehensive data set, wherein the spatial analysis model includes a terrain analysis model, a road network accessibility model, a land use suitability model, and a population density distribution model; Step S24: generating candidate locations for the spatial analysis model and the charging demand distribution map data according to a preset location constraint condition set, thereby obtaining a preliminary candidate location set; Step S25: Perform a comprehensive evaluation of the preliminary candidate location set based on accessibility, power grid capacity, and surrounding facilities to obtain comprehensive scoring data for the candidate locations; Step S26: Perform spatial cluster analysis on the candidate location comprehensive score data and perform location suitability scoring to obtain location suitability scoring data.

[0043] As an embodiment of the present invention, refer to Figure 3 As shown, Figure 1 Detailed step flow diagram of step S2 in the embodiment of the present invention, step S2 includes the following steps: Step S21: Acquire geographic information data of the target area, including terrain data, road network data, land use data, population density data, existing charging facility distribution data, and power grid distribution data; The embodiments of the present invention obtain elevation data, topographic maps, etc. of the target area, which are usually obtained from satellite images or geographic information system (GIS) databases, obtain vector data of the road network, including the grade, width and flow information of the road, collect data on land use types, such as residential areas, commercial areas, industrial areas, etc., which are usually obtained from urban planning departments, obtain population density data of different areas, which may include statistical data or census data released by the Statistics Bureau, obtain the location and quantity data of existing charging piles in the target area, which can be obtained from charging network providers or public data sources, obtain infrastructure data of the power grid, including the location of substations, power line layout, etc., and obtain data from public data platforms, government statistics bureaus, GIS databases, field surveys and other channels.

[0044] Step S22: preprocessing and integrating the geographic information data to obtain a comprehensive geographic information data set, wherein the data preprocessing includes data cleaning, format unification, and coordinate system conversion; The embodiments of the present invention identify and remove errors, duplicate records and irrelevant information in the data, interpolate or delete missing data to ensure data integrity; convert data from different sources into a unified format (such as GeoJSON, Shapefile) for easy integration and analysis; unify all data into the same geographic coordinate system to ensure accurate alignment of spatial data. The cleaned and formatted data is merged into a comprehensive data set to ensure that all relevant data can be used in the same model.

[0045] Step S23: constructing a spatial analysis model based on the geographic information comprehensive data set, wherein the spatial analysis model includes a terrain analysis model, a road network accessibility model, a land use suitability model, and a population density distribution model; The embodiment of the present invention uses terrain data to create a DEM to analyze the terrain slope and height, and extract the impact of terrain features such as mountains and rivers on the layout of charging piles. Use road network data to create a road network map, and calculate the accessibility of different locations to major road nodes through the shortest path algorithm (such as the Dijkstra algorithm). Based on land use data, define the suitability of different land types for the layout of charging piles, and score them according to the suitability of land use types (for example, commercial areas have high suitability and industrial areas have low suitability). Use population density data to create a population distribution map to identify densely populated areas, which may require more charging piles.

[0046] Step S24: generating candidate locations for the spatial analysis model and the charging demand distribution map data according to a preset location constraint condition set, thereby obtaining a preliminary candidate location set; The embodiment of the present invention defines location constraints, environmental constraints such as areas where charging piles are not allowed to be set up, such as protected areas and historical sites, and regulatory requirements such as minimum distance from buildings, noise restrictions, etc. The spatial analysis model is combined with the charging demand distribution map data, and the location constraints are considered to generate charging pile candidate locations; potential charging pile locations are screened out from the areas that meet the conditions to generate a preliminary candidate location set.

[0047] Step S25: Perform a comprehensive evaluation of the preliminary candidate location set based on accessibility, power grid capacity, and surrounding facilities to obtain comprehensive scoring data for the candidate locations; The embodiment of the present invention scores the candidate location based on the distance from the main roads and public transportation, takes into account traffic flow and accessibility, checks the grid load capacity of each candidate location, evaluates whether it can meet the power demand of the charging pile, analyzes the density of commercial centers, public facilities and residential areas around the candidate location, evaluates the impact on the use of the charging pile, assigns weights according to the importance of different factors (such as 40% for accessibility, 30% for grid capacity, and 30% for surrounding facilities), comprehensively scores the candidate location, and obtains comprehensive scoring data for the candidate location.

[0048] Step S26: Perform spatial cluster analysis on the candidate location comprehensive score data and perform location suitability scoring to obtain location suitability scoring data.

[0049] The embodiment of the present invention uses a spatial clustering algorithm (such as K-means, DBSCAN) to perform cluster analysis on the comprehensive score data of candidate locations, and identifies areas with higher comprehensive scores as charging demand hotspots. The locations in the clustering results are scored for suitability to evaluate whether these areas meet the requirements for setting up charging piles; the spatial clustering results and the comprehensive scores are combined to obtain the final candidate location suitability score data.

[0050] By acquiring a variety of geographic information data, the present invention can fully understand the geographical environment and infrastructure conditions of the target area, and provide rich data information for subsequent analysis; different types of data (such as terrain, roads, population density, etc.) provide support from multiple dimensions to ensure the scientificity and rationality of the planning scheme. Through data preprocessing, inaccurate or incomplete data are cleaned, the data format and coordinate system are unified, and the quality and consistency of the data are improved; data from different sources are integrated to form a comprehensive data set to provide a basis for subsequent spatial analysis. Construct multiple spatial analysis models to analyze the characteristics of the target area from different angles to provide comprehensive information support for the site selection of charging piles; through models such as terrain, road network, land use and population density, regional characteristics can be refined to help identify locations suitable for installing charging piles. According to the preset constraints, ensure that the candidate locations meet the actual needs and restrictions, and improve the feasibility and effectiveness of the planning scheme; generate a preliminary candidate location set, conduct preliminary screening, and lay the foundation for subsequent comprehensive evaluation. Through a multi-factor comprehensive evaluation of the candidate locations, multiple factors such as accessibility, power grid capacity and surrounding facilities are considered to ensure the comprehensiveness and rationality of the site selection; generate comprehensive scoring data for the candidate locations to provide a scientific basis for the final site selection decision. Through spatial cluster analysis, we can identify candidate locations that are geographically close and have high scores, optimize the site selection layout, and score the suitability of the locations to provide clear site selection suggestions, thereby improving the scientificity and reliability of charging pile planning. Through the above steps, we can systematically collect, preprocess and analyze the geographic information of the target area, build a multi-dimensional spatial analysis model, and conduct a comprehensive evaluation and suitability scoring of candidate locations from multiple perspectives. These steps not only improve the quality and consistency of the data, but also ensure the scientificity, rationality and feasibility of the site selection through comprehensive evaluation of multiple factors and spatial cluster analysis, providing comprehensive and scientific decision-making support for charging pile planning.

[0051] Preferably, step S251 includes the following steps: Step S251: Calculate the shortest path and travel time from the candidate location to the main road node for the preliminary candidate location set of the road network data, and perform weighted calculation based on the preset accessibility weight to obtain accessibility score data; The embodiment of the present invention imports the road network data into a geographic information system (GIS) or a dedicated path analysis tool, and uses a shortest path algorithm (such as a Dijkstra algorithm or an A* algorithm) to calculate the shortest path from each candidate location to a major road node (such as a major intersection, a major road); based on the traffic flow and speed limit of the road network, the travel time from each candidate location to the major road node is calculated. The shortest path and travel time are weighted according to a preset weight, and the weight can be based on factors such as the level of traffic flow and the type of road; the weighted shortest path and travel time data are integrated to obtain the accessibility score data of each candidate location.

[0052] Step S252: obtaining the point of interest data of the target area, wherein the points of interest include commercial centers, public facilities and residential areas; performing density and type statistics of the points of interest within a certain range around each candidate location based on the point of interest data and the preliminary candidate location set, thereby obtaining the original data of the surrounding facilities; The embodiment of the present invention obtains point of interest data from a map service provider, local government or database, including commercial centers, public facilities (such as hospitals, schools), residential areas, etc.; ensures that the point of interest data has geographic coordinate information to facilitate spatial analysis in GIS. Define a surrounding buffer zone (such as a radius of 500 meters or 1 kilometer) for each candidate location, count the number of different types of points of interest in the buffer zone, calculate the density of the points of interest (the number of points of interest per square kilometer), record the types and quantities of different types of points of interest, and obtain the original data of surrounding facilities.

[0053] Step S253: weighting and scoring the original data of surrounding facilities according to the preset facility type weights, thereby obtaining surrounding facility scoring data; The embodiment of the present invention sets weights according to the impact of different types of facilities on the layout of charging piles. For example, commercial centers and public facilities may be more important than residential areas, and higher weights may be assigned. The preset weights are applied to the original data of surrounding facilities, and the weighted surrounding facility scores of each candidate location are calculated. The weighted surrounding facility score data are integrated to obtain the final surrounding facility score data of each candidate location.

[0054] Step S254: Calculate the distance from each candidate location to the nearest substation and the surrounding power grid capacity according to the power grid distribution data, thereby obtaining the original data of the power grid capacity; and perform weighted calculation on the original data of the power grid capacity according to the preset power grid capacity weight, thereby obtaining the power grid capacity scoring data; The embodiment of the present invention extracts the geographical location of the substation from the power grid distribution data, uses a spatial analysis tool to calculate the distance from each candidate location to the nearest substation; extracts the capacity information of the power grid around each candidate location, including the capacity of the substation and the power line, and evaluates the power grid load capacity of each candidate location based on the power grid capacity data. Set weights for the distance from the substation and the power grid capacity to reflect the relative importance of the power grid carrying capacity; weight the original power grid capacity data according to the weights, and calculate the power grid capacity score data for each candidate location.

[0055] Step S255: Perform a multi-factor comprehensive scoring based on the accessibility scoring data, the surrounding facilities scoring data, the power grid capacity scoring data, and the charging demand distribution map data, so as to obtain the candidate location comprehensive scoring data.

[0056] The embodiment of the present invention combines the accessibility score, the surrounding facility score and the grid capacity score data, and considers the charging demand distribution map data; weights are set according to the importance of different scoring factors. For example, accessibility may be more important than grid capacity and can be given a higher weight; the weights are applied to perform weighted summation on each score data to obtain the comprehensive score data for each candidate location. According to the calculated comprehensive score data, the candidate locations are sorted and screened, and the candidate locations with higher comprehensive scores are selected as the optimal layout points of the charging piles.

[0057] The present invention can evaluate the traffic convenience of the candidate location by calculating the shortest path and driving time from the candidate location to the main road node, and provide convenient charging services for users; weighted calculation is performed based on the preset accessibility weight, so that the accessibility score is more scientific and reasonable, reflecting the differences in traffic convenience of different locations; through the accessibility score, the location with convenient traffic can be effectively screened out, and the utilization rate and user satisfaction of charging piles can be improved. Through the density and type statistics of the point of interest data, the commercial, public facilities and residential areas around the candidate location can be fully understood, and a convenient charging environment can be provided for users; high-density and diversified points of interest can attract more users to choose these charging pile locations, thereby improving user experience and the utilization rate of charging piles. Weighted scoring is performed according to the weights of different types of facilities, and the surrounding facilities of the candidate location can be scientifically evaluated to ensure the objectivity and accuracy of the scoring results; through the surrounding facility scoring, the candidate location with complete supporting facilities can be screened out, and convenient charging services can be provided to users, which enhances the attractiveness of charging piles. By calculating the distance from the candidate location to the nearest substation and the surrounding power grid capacity, the reliability and feasibility of power supply can be evaluated to ensure the stable operation of the charging pile; weighted calculation based on the power grid capacity weight can comprehensively consider various factors of power supply, making the scoring result more scientific and reasonable; through the power grid capacity scoring, candidate locations with sufficient power supply and moderate distance from substations can be screened out, power resource allocation can be optimized, and efficient operation of charging piles can be ensured. Comprehensively considering accessibility, surrounding facilities, power grid capacity, and charging demand distribution, the advantages and disadvantages of candidate locations can be comprehensively evaluated to ensure that the selected charging pile location has all-round advantages; through multi-factor comprehensive scoring, the best candidate charging pile location can be scientifically screened to improve the utilization rate and service quality of charging piles; comprehensive scoring data provides data-driven decision support for charging pile planning to ensure the scientificity and feasibility of the planning scheme.

[0058] Preferably, step S255 includes the following steps: Step S2551: performing data alignment based on the preliminary candidate location set according to the accessibility score data, the surrounding facility score data, and the power grid capacity score data, and performing standardization processing to obtain a standardized score data set; The embodiment of the present invention integrates the accessibility score data, the surrounding facility score data and the power grid capacity score data into a unified database for subsequent processing; ensures that all score data correspond to the same candidate location, and ensures that the score data for each location is complete. Select a suitable standardization method, such as Z-Score standardization (converting the data to a distribution with a mean of 0 and a standard deviation of 1) or minimum-maximum standardization (scaling the data to between 0 and 1); standardize the accessibility score, surrounding facility score and power grid capacity score data to obtain a standardized score data set. This process eliminates the influence of different score data dimensions, so that data of different score dimensions can be compared fairly.

[0059] Step S2552: performing weighted summation on the standardized scoring data set to obtain preliminary comprehensive scoring data; The embodiment of the present invention defines the weights of the accessibility score, the surrounding facility score, and the power grid capacity score according to the importance of each scoring dimension in the overall evaluation. For example, the accessibility score may be more important, so a higher weight can be assigned. Multiply each scoring data in the standardized scoring data set by its corresponding weight, and add all weighted results to obtain preliminary comprehensive scoring data; comprehensive score = weight 1 × standardized accessibility score + weight 2 × standardized surrounding facility score + weight 3 × standardized power grid capacity score.

[0060] Step S2553: performing spatial autocorrelation analysis based on the spatial relationship between candidate positions according to the preliminary comprehensive scoring data, and performing spatial smoothing processing to obtain comprehensive scoring data; The embodiment of the present invention uses a spatial autocorrelation analysis method (such as Moran's I index) to evaluate the spatial autocorrelation of the comprehensive score of the candidate location, that is, whether the distribution of the score value in space is regular; calculate the correlation between the comprehensive score of each location and the comprehensive score of its neighboring locations to identify clustered areas with high or low scores. Select a suitable spatial smoothing method (such as interpolation, kernel density estimation or moving average) to smooth the comprehensive score data, smooth the comprehensive score data, eliminate the extreme values ​​of the local score, enhance the spatial continuity of the score data, and obtain the smoothed comprehensive score data.

[0061] Step S2554: Perform demand-weighted adjustment on the comprehensive score data according to the charging demand distribution map data, so as to obtain comprehensive score data of the candidate locations.

[0062] The embodiment of the present invention extracts the charging demand data of each candidate location from the charging demand distribution map, and defines a weighting factor for each candidate location according to the intensity of the charging demand. Areas with high demand can be given higher weights. The demand weighting factor is applied to the comprehensive score data to adjust the comprehensive score. The adjustment can be completed by multiplying the comprehensive score by the demand weighting factor; the weighted candidate location comprehensive score data is obtained, which reflects the combination of the comprehensive score and the actual charging demand.

[0063] The data alignment and standardization processing of the present invention ensure that different scoring data (accessibility, surrounding facilities, power grid capacity) are compared under the same scale and unit, avoiding deviations caused by different dimensions; standardization processing makes the impact of various scoring indicators on the final comprehensive score more balanced, avoiding excessive impact of a certain factor on the final score due to dimensional differences; the data after standardization processing is more accurate and comparable, which helps to improve the scientificity and accuracy of subsequent comprehensive scores. Weighted summation can comprehensively consider the accessibility score, surrounding facility score and power grid capacity score to obtain a comprehensive scoring data, so that the advantages and disadvantages of candidate locations can be more comprehensively evaluated; through weighted processing, the importance of different scoring indicators in the comprehensive score can be reflected, so that the scoring results are more in line with actual needs and planning goals; the preliminary comprehensive scoring data provides a basis for further screening of candidate locations and helps identify relatively superior locations. Spatial autocorrelation analysis can identify the spatial correlation between candidate locations and help understand the relative advantages and disadvantages of the locations and the spatial distribution pattern; spatial smoothing can reduce random fluctuations in the scoring data, making the scoring results smoother and more stable, and helping to more accurately reflect the actual advantages and disadvantages of each candidate location; through spatial autocorrelation analysis and spatial smoothing, the spatial layout of charging piles can be better optimized, avoiding concentrated settings or blank areas, and improving the overall service efficiency of charging piles. According to the charging demand distribution map data, weighted adjustment is made to make the final comprehensive score more in line with the actual charging demand, ensuring that the layout of charging piles can meet the actual needs of users; through demand weighted adjustment, charging piles can be set in areas with higher demand, thereby improving the coverage and quality of charging services; the final comprehensive score data can more accurately reflect the charging demand, optimize the configuration of charging pile resources, improve resource utilization efficiency, and ensure the scientificity and rationality of charging pile settings. These steps combine data standardization, weighted summation, spatial autocorrelation analysis, and demand weighted adjustment to ensure that the comprehensive score of candidate locations is both comprehensive and accurate. Through such processing, the setting location of charging piles can be effectively optimized, the coverage and quality of charging services can be improved, and the scientificity and rationality of resource allocation can be improved. This method can provide more accurate data support for charging pile planning, ensure that the planning scheme is closer to actual needs, and ultimately improve the efficiency of charging services and user satisfaction.

[0064] Preferably, step S3 comprises the following steps: Step S31: Acquire technical parameter data of different types of charging piles, including charging power, charging efficiency and interface type; The embodiment of the present invention defines the required technical parameters, including charging power (for example, power specifications of AC slow charging, DC fast charging, etc.), charging efficiency (conversion efficiency of electric energy during charging), interface type (for example, Type 1, Type 2, CCS, etc.); collect technical parameter data from different charging pile manufacturers or suppliers. This information can be obtained through product manuals, technical specification sheets or data provided by suppliers. The collected technical parameter data is organized into a database form to ensure that the data is accurate and easy to retrieve; all technical parameter data are converted into a unified format for subsequent analysis and processing.

[0065] Step S32: Obtain cost data of different types of charging piles, including equipment cost, installation cost, operation cost and maintenance cost; merge the technical parameter data and cost data into technical parameter and cost data; The embodiment of the present invention specifies the equipment cost (cost of purchasing the charging pile), installation cost (cost required to install the charging pile), operating cost (energy consumption and management cost during operation), and maintenance cost (fault repair and daily maintenance cost), and obtains relevant cost data from the charging pile supplier or market research report. The technical parameter data collected in step S31 is combined with the cost data collected in step S32 to create a comprehensive database containing charging pile technical parameters and cost information; ensure that all data are accurate and relevant to facilitate subsequent modeling and analysis.

[0066] Step S33: establishing a charging pile performance-cost model according to the technical parameters and cost data, thereby obtaining charging pile performance-cost evaluation data; The embodiment of the present invention constructs a relationship model between charging pile performance and cost. Generally, this includes establishing a linear or nonlinear regression model to describe the relationship between technical parameters (such as power and efficiency) and cost; using statistical software or programming tools (such as Python, R) to perform data analysis, fit the performance-cost model, and analyze the impact of different technical parameters on cost. Use historical data or test data to verify the accuracy of the model to ensure that the model can reasonably predict the performance and cost of different charging piles; calculate the performance-cost evaluation data of different charging piles based on the model to provide basic data for multi-objective optimization.

[0067] Step S34: setting an objective function of multi-objective optimization, including minimizing total cost, maximizing coverage, and maximizing service quality; determining the potential demand of each candidate location according to the location suitability score data and the charging demand distribution map data, thereby obtaining potential demand data; The embodiment of the present invention defines the objective function of multi-objective optimization, including minimizing the total cost (the sum of equipment, installation, operation, and maintenance costs), maximizing the coverage (the geographical scope of the charging pile service), and maximizing the service quality (such as the reliability of the charging pile, charging speed, etc.). Based on the location suitability score data and the charging demand distribution map data, the potential charging demand for each candidate location is calculated, and the demand can be estimated using statistical analysis methods or prediction models, and the potential demand data is combined with the charging pile performance-cost data to form a data set for optimization analysis.

[0068] Step S35: Solve the objective function according to the charging pile performance-cost evaluation data and the potential demand data to obtain a non-dominated solution set; evaluate and screen the non-dominated solution set to obtain a preliminary configuration plan.

[0069] The embodiment of the present invention selects a suitable multi-objective optimization algorithm, such as genetic algorithm, particle swarm optimization or simulated annealing, etc.; inputs charging pile performance-cost evaluation data and potential demand data, solves the objective function, and obtains a set of non-dominated solutions (i.e., the solution with the best performance among all objectives). The non-dominated solution set is evaluated, the performance of each solution under different objectives is analyzed, and the trade-off factors in practical applications are considered to screen out the optimal solution; based on the screening results, a preliminary configuration plan is generated to determine the type, quantity and location of the charging piles.

[0070] The present invention obtains the technical parameter data of the charging pile, which is helpful to understand the performance characteristics of different charging piles, such as charging power and efficiency, which are essential for selecting the appropriate type and specification of charging piles; it can compare the technical indicators of different types of charging piles, select the charging pile that best suits the charging needs of the target area, and thus optimize the configuration scheme of the charging piles. By collecting and merging equipment costs, installation costs, operating costs, and maintenance costs, the total cost of different types of charging piles can be fully understood. This helps to evaluate the economic efficiency of the charging piles; integrating technical parameters with cost data helps to evaluate the cost-effectiveness of different charging piles in different usage scenarios, so as to select the most cost-effective type of charging pile. Establishing a performance-cost model can find a balance between technical performance and economic cost, which helps to select the best type of charging pile in terms of performance and cost; performance-cost evaluation data provides a quantitative basis for decision-making, helping to find the most suitable solution among multiple charging pile selection schemes. Setting a multi-objective optimization objective function can simultaneously consider cost, coverage, and service quality, so as to ensure that the comprehensive benefits of the charging pile configuration scheme are maximized; through the analysis of potential demand data, the charging demand of each candidate location can be accurately understood, so as to better meet the actual needs in the optimization process and improve the coverage and quality of charging services. By solving the objective function, a non-dominated solution set (i.e., a solution that is not completely dominated by other solutions) is obtained, which can provide multiple effective configuration solutions for selection; the non-dominated solution set allows decision makers to choose from a variety of optimization solutions, comprehensively consider the balance of cost, coverage, and service quality, and obtain the most suitable preliminary configuration solution; through further evaluation and screening of the non-dominated solution set, a configuration solution that is more in line with actual needs and goals can be obtained, improving the scientificity and effectiveness of the configuration solution. These steps ensure that the configuration solution of charging piles is both economical and practical, and can meet the actual charging needs of the region by combining technical parameters, cost data, multi-objective optimization, and demand analysis. The performance-cost model provides a scientific evaluation basis, multi-objective optimization ensures the maximization of comprehensive benefits, and the generation and screening of the non-dominated solution set provides a flexible basis for decision-making. These steps work together to develop the optimal configuration solution for charging piles, improve the efficiency and quality of charging services, and optimize the allocation of resources.

[0071] Preferably, step S4 comprises the following steps: Step S41: Acquire power grid information data of the target area, including substation distribution, power line layout and load distribution; The embodiment of the present invention obtains the geographic location and capacity data of the substation from the power company or the grid operator. The data may include the specific location of the substation, the capacity of the transformer, the service range, etc.; collects the layout information of the power lines, including the direction, capacity, structure, etc. of the transmission lines and distribution lines, which can usually be obtained from the map or data system provided by the power company; obtains the load distribution data of the target area, including the power demand situation in different areas, which can be obtained through historical load data, forecast data or actual measurement data. The substation distribution, power line layout and load distribution data are integrated into a comprehensive database for subsequent analysis, ensuring that all data formats are unified for easy processing and analysis.

[0072] Step S42: constructing a power system capacity model according to the power grid information data, wherein constructing the power system capacity model specifically includes constructing a power grid topology model describing node and line information according to the power grid information data, establishing a line capacity limitation model according to the thermal capacity and stability limitation of the line, constructing a substation capacity limitation model based on the capacity constraint of the transformer and the switchgear, and integrating the power grid topology model, the line capacity limitation model and the substation capacity limitation model; The embodiment of the present invention establishes a power grid topology model that describes power grid nodes (such as substations, distribution boxes) and lines (such as transmission lines, distribution lines) based on power grid information data. The modeling can be performed using power system modeling software or programming tools; the connection relationship between nodes is determined, and the capacity, length and other parameters of each line are defined. According to the design parameters and actual load conditions of the line, a line capacity limitation model is established, taking into account the thermal capacity (maximum allowable load) of the line and the stability limitation of the line (to avoid overload and overtemperature). According to the equipment capacity of the substation (such as the capacity of transformers and switchgear), a substation capacity limitation model is established to ensure that the substation does not exceed its carrying capacity under various load conditions. The power grid topology model, the line capacity limitation model and the substation capacity limitation model are integrated to form a comprehensive power system capacity model.

[0073] Step S43: Analyze the grid carrying capacity of the preliminary configuration plan according to the power system capacity model to obtain grid load impact assessment data; The embodiment of the present invention inputs the preliminary configuration plan (including the location and type of charging piles) into the power system capacity model, calculates the load of each grid node and line according to the number and location of charging piles in the configuration plan, and evaluates whether the load of each grid node and line under the configuration of charging piles exceeds its carrying capacity. Check whether the load of the substation is within its capacity limit, generate grid load impact assessment data, and indicate which parts of the grid may be overloaded or at risk.

[0074] Step S44: According to the grid load impact assessment data, the charging piles that do not meet the grid carrying capacity requirements are adjusted, and feasibility verification is performed to obtain an adjusted feasible configuration plan; The embodiment of the present invention adjusts the configuration of charging piles that do not meet the carrying capacity requirements based on the grid load impact assessment data, which may include reducing the number of charging piles in certain areas, changing the type of charging piles, or reselecting locations. The adjusted charging pile configuration scheme is input into the power system capacity model, and the carrying capacity analysis is re-performed; check whether the adjusted configuration scheme meets the carrying capacity requirements of the grid, and if so, generate an adjusted feasible configuration scheme.

[0075] Step S45: Simulate the operation status of the charging station in different time periods and scenarios according to the feasible configuration scheme, and perform performance evaluation based on the utilization rate, waiting time and grid load to obtain performance evaluation data.

[0076] The embodiment of the present invention sets the operation status of the charging station under different time periods (such as peak period and non-peak period) and scenarios (such as different charging demand modes); uses the charging station operation simulation software or programming tools to perform operation simulation under different scenarios, and generates data such as the utilization rate and waiting time of the charging station. Analyze the utilization rate of the charging station in different time periods and check the utilization efficiency of the charging piles; evaluate the waiting time of users at the charging station to ensure that the user experience meets the requirements; check the grid load in different scenarios to ensure that the grid can operate stably during the peak charging period, generate performance evaluation data based on the utilization rate, waiting time and grid load, and provide the overall performance of the charging station configuration.

[0077] The present invention obtains power grid information data to fully understand the power infrastructure of the target area, including substations, power lines and load distribution, which helps to evaluate the current status and capacity of the power grid; it provides the necessary basic data for building a power system capacity model to ensure the accuracy and reliability of subsequent analysis. Building a power system capacity model can comprehensively evaluate the carrying capacity of the power grid, including power grid topology, line capacity and substation capacity, which helps to ensure that the power grid can support the configuration of charging piles; by establishing a detailed line capacity limitation model and a substation capacity limitation model, it is possible to gain an in-depth understanding of the specific constraints of the power grid and avoid operational problems caused by power grid overload; model integration can provide a global power grid capacity assessment framework to ensure coordination and comprehensive capacity analysis between various power grid components. Performing a power grid carrying capacity analysis on the preliminary configuration plan can evaluate the impact of the charging pile configuration on the power grid load, which helps identify potential power grid pressure points and ensures that the configuration plan does not exceed the load capacity of the power grid; it can promptly discover and solve problems with the power grid load to avoid unstable power supply caused by excessive power grid load. Adjusting the charging piles that do not meet the grid carrying capacity requirements can optimize the configuration of the charging piles so that they can operate within the grid capacity. Feasibility verification is performed to ensure that the adjusted configuration plan can operate effectively under actual conditions and reduce the risk of problems after implementation. By simulating the operation of charging stations in different time periods and scenarios, the performance of charging stations under various actual conditions can be evaluated, which helps to understand the performance of charging pile configuration in actual operation. Performance evaluation based on utilization rate, waiting time and grid load can comprehensively analyze the service quality and grid load of charging piles to ensure that charging stations do not put too much pressure on the grid while meeting user needs. Performance evaluation data provides data support for further optimization of configuration plans, helps improve the layout and operation strategy of charging piles, and improves the overall efficiency of the system. These steps ensure that the charging pile configuration plan does not put too much burden on the grid while meeting charging needs by comprehensively considering the mutual influence of grid information and charging pile configuration. The constructed power system capacity model and load impact assessment can provide accurate grid carrying capacity analysis, and the adjustment and verification steps ensure the feasibility and stability of the final plan. Dynamic simulation and performance evaluation provide detailed reference data for actual operation, thereby optimizing the configuration and operation strategy of charging piles and improving the efficiency and reliability of the overall system.

[0078] Preferably, step S5 comprises the following steps: Step S51: identifying the locations and time periods of charging stations that do not meet the preset performance targets according to the performance evaluation data, thereby obtaining performance target difference data; The embodiment of the present invention determines preset performance targets, such as the utilization rate of charging stations should reach 90%, the waiting time should be less than 15 minutes, the grid load should not exceed 80%, etc.; obtains performance evaluation data from the previous steps, including the utilization rate, waiting time, and grid load data of charging stations; compares these performance evaluation data with the preset performance targets to identify which charging stations and time periods fail to meet the targets. Based on the comparison results, the locations and time periods of charging stations that do not meet the preset performance targets are marked, and these data will help identify specific areas and time periods that need to be optimized; the locations, time periods, and related differences of charging stations that do not meet the performance targets are organized into performance target difference data, which will serve as the basis for subsequent optimization.

[0079] Step S52: taking the performance target as a constraint condition according to the preset performance target and the performance target difference data, and taking minimizing the adjustment cost as the objective function, thereby establishing a performance dynamic optimization model; The embodiment of the present invention uses the performance target difference data as a constraint condition to define the constraints in the optimization model. For example, the utilization rate of the charging station after adjustment must reach 90%, and the waiting time must be less than 15 minutes, etc.; the objective function is set to minimize the adjustment cost. This includes the equipment cost, installation cost, operating cost and maintenance cost of the charging pile, and a formula can be used to represent the calculation of the adjustment cost. Establish a mathematical optimization model in which the objective function is to minimize the adjustment cost, and the constraints include all restrictions in the performance target difference data; select a suitable optimization algorithm, such as linear programming, integer programming or nonlinear programming, to solve this optimization model, ensure that the model can correctly represent the actual problem, and check whether all constraints and objectives are correctly incorporated into the model.

[0080] Step S53: Design an optimization algorithm based on the adjustment of the number of charging piles, the replacement of the type, and the fine-tuning of the position according to the performance dynamic optimization model, and optimize the feasible configuration scheme using the optimization algorithm to obtain the final dynamic planning scheme.

[0081] The embodiment of the present invention selects a suitable optimization algorithm. According to the complexity of the optimization model, genetic algorithms, particle swarm optimization, simulated annealing and other algorithms can be used; the algorithm is designed to handle operations such as charging pile quantity adjustment, type change, and position fine-tuning. The algorithm should be able to handle different types of variables and constraints. The data of the performance dynamic optimization model is input into the optimization algorithm, and the optimization algorithm is run for calculation to obtain the optimized charging pile configuration plan; according to the algorithm results, the number, type and location of the charging piles are adjusted to minimize the adjustment cost and meet the performance goals. Verify whether the optimized configuration plan meets all preset performance goals; generate the final dynamic planning plan, including information such as the adjusted charging pile configuration, type and location. According to the final dynamic planning plan, the actual charging pile adjustment and deployment are carried out; after implementation, the performance of the charging station is continuously monitored to ensure the effectiveness of the optimization plan, and further adjustments are made according to actual conditions.

[0082] The present invention can accurately locate the deficiencies in the system by identifying the locations and time periods of charging stations that do not meet the preset performance targets, which helps to understand in which areas or time periods the performance of the charging piles does not meet expectations; the obtained performance target difference data provides a specific improvement direction for subsequent optimization, ensuring that the optimization process is targeted and can solve practical problems. By taking the performance target as a constraint, the optimization model clarifies the optimization direction, that is, while meeting the performance target, the adjustment cost is minimized, which helps to balance the relationship between performance improvement and cost control; the objective function of minimizing the adjustment cost ensures that when adjusting the scheme, not only performance improvement but also cost-effectiveness is focused on, thereby improving the optimization efficiency and economy. By adjusting the number, type and location of charging piles through the optimization algorithm, a more reasonable configuration can be achieved, and the performance of the overall system can be improved, which ensures that the final solution can achieve the best balance between cost and performance; the optimization algorithm can handle different types of adjustments and provide flexible solutions, so that the final planning scheme can adapt to changes in actual needs; the final optimization of the dynamic planning scheme enables the configuration scheme to be adjusted in time according to the actual usage and the results of performance evaluation, thereby enhancing the adaptability and stability of the system. These steps ensure that the charging pile configuration scheme reduces the adjustment cost as much as possible while meeting the preset performance target by identifying and analyzing the performance target difference data. By establishing a dynamic performance optimization model and designing an optimization algorithm, the final planning scheme can not only improve the service quality of the charging pile, but also achieve the best cost control. Such a dynamic adjustment process can make the charging pile system more efficient, flexible and economical, and ultimately improve the overall user experience and system stability.

[0083] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0084] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A dynamic analysis charging pile planning method, characterized in that: The following steps are involved: Step S1: acquiring electric vehicle usage data in the target area, performing charging demand forecasting analysis based on the electric vehicle usage data, and obtaining a charging demand forecasting model; performing multi-scenario simulation based on the charging demand forecasting model, and obtaining charging demand distribution map data, wherein the multi-scenario includes scenarios with different traffic flows, different driving modes, and different charging demands; Step S2: Acquire geographic information data of the target area; A spatial analysis model is constructed based on geographic information data, and candidate locations for charging piles are evaluated based on the charging demand distribution map data to obtain location suitability score data; Step S3: Establish a charging pile performance-cost model based on the pre-acquired technical parameters and cost data of different types of charging piles, and perform a multi-objective optimization analysis based on cost-coverage-service quality based on the location suitability score data to obtain a preliminary configuration plan; Step S4: constructing a power system capacity model according to the power grid information of the target area; analyzing the power grid carrying capacity of the preliminary configuration plan according to the power system capacity model, and adjusting the configuration plan to obtain a feasible configuration plan; Dynamically simulate feasible configuration solutions and conduct performance evaluation based on utilization rate, waiting time and grid load to obtain performance evaluation data; Step S5: According to the preset performance target and performance evaluation data, the feasible configuration scheme is dynamically optimized and adjusted driven by the performance target, so as to obtain the final dynamic planning scheme.

2. The charging pile planning method of dynamic analysis according to claim 1 is characterized in that: Step S11 includes the following steps: Step S11: Acquire electric vehicle usage data in the target area, including vehicle type data, driving mode data, and charging demand data; Step S12: standardizing the electric vehicle usage data and processing outliers to obtain electric vehicle standard data; Step S13: extracting driving mode features based on average driving distance and frequency from the electric vehicle standard data, thereby obtaining driving mode feature data; Step S14: extracting charging behavior characteristics based on charging frequency, charging amount, charging type and charging time from the electric vehicle standard data, thereby obtaining charging behavior characteristic data; Step S15: clustering the driving mode characteristic data and the charging behavior characteristic data into different types of vehicle characteristics according to the vehicle type data, thereby obtaining vehicle type usage characteristic data; Step S16: performing charging demand forecasting analysis based on the driving mode characteristic data, the charging behavior characteristic data, and the vehicle type usage characteristic data to obtain a charging demand forecasting model; Step S17: Perform multi-scenario simulation based on the charging demand prediction model to obtain charging demand distribution map data, wherein the multi-scenario includes scenarios with different traffic flows, different driving modes, and different charging demands.

3. The method for planning charging piles by dynamic analysis according to claim 2, characterized in that: Step S16 includes the following steps: Step S161: merging the driving mode characteristic data, the charging behavior characteristic data and the vehicle type usage characteristic data into a data set, and performing a data integrity check and missing value processing to obtain a comprehensive characteristic data set; Step S162: performing time series processing on the comprehensive feature data set, and performing trend, seasonality and random component decomposition, thereby obtaining comprehensive feature time series data; Step S163: Calculate the correlation coefficient of the comprehensive characteristic time series data and the charging demand data to obtain correlation coefficient data, wherein the correlation coefficient calculation includes calculating the linear correlation between continuous variables by using the Pearson correlation coefficient, calculating the nonlinear correlation of categorical variables by using the Spearman rank correlation coefficient, and calculating the time lag correlation by using the cross correlation function; Step S164: performing feature selection based on Gini impurity on the comprehensive feature data set according to the correlation coefficient data, thereby obtaining a demand key feature set; Step S165: Using the demand key feature set as input features and the charging demand data as the target variable, constructing and training a decision tree model, and optimizing the model parameters through cross-validation to obtain a charging demand prediction model.

4. The method for planning charging piles by dynamic analysis according to claim 3, characterized in that: Step S17 includes the following steps: Step S171: Design multiple scenarios based on different traffic flows, different driving modes, and different charging demands according to the charging demand prediction model, and parameterize the scenarios to obtain scenario parameter setting data, wherein different traffic flows include increased charging demand during the morning peak, increased charging demand during the evening peak, and uniform charging demand throughout the day; different driving modes include mainly short-distance driving within the city, mixed driving in the city and suburbs, and mainly long-distance high-speed driving; different charging demands include mainly charging at home at night, mainly charging at public charging piles, and mixed charging demands; Step S172: performing Monte Carlo simulation on the scenario parameter setting data to generate charging scenario simulation data; Step S173: inputting the charging scenario simulation data into the charging demand prediction model to obtain charging demand prediction result data under different scenarios; Step S174: Perform a charging demand change analysis based on the time dimension on the charging demand prediction result data, perform a spatial clustering analysis based on the charging demand quantity, and identify charging demand hot spots, thereby obtaining charging demand distribution map data.

5. The method for planning charging piles by dynamic analysis according to claim 4, characterized in that: Step S2 includes the following steps: Step S21: Acquire geographic information data of the target area, including terrain data, road network data, land use data, population density data, existing charging facility distribution data, and power grid distribution data; Step S22: preprocessing and integrating the geographic information data to obtain a comprehensive geographic information data set, wherein the data preprocessing includes data cleaning, format unification, and coordinate system conversion; Step S23: constructing a spatial analysis model based on the geographic information comprehensive data set, wherein the spatial analysis model includes a terrain analysis model, a road network accessibility model, a land use suitability model, and a population density distribution model; Step S24: generating candidate locations for the spatial analysis model and the charging demand distribution map data according to a preset location constraint condition set, thereby obtaining a preliminary candidate location set; Step S25: Perform a comprehensive evaluation of the preliminary candidate location set based on accessibility, power grid capacity, and surrounding facilities to obtain comprehensive scoring data for the candidate locations; Step S26: Perform spatial cluster analysis on the candidate location comprehensive score data and perform location suitability scoring to obtain location suitability scoring data.

6. The method for planning charging piles by dynamic analysis according to claim 5, characterized in that: Step S251 includes the following steps: Step S251: Calculate the shortest path and travel time from the candidate location to the main road node for the preliminary candidate location set of the road network data, and perform weighted calculation based on the preset accessibility weight to obtain accessibility score data; Step S252: obtaining the point of interest data of the target area, wherein the points of interest include commercial centers, public facilities and residential areas; performing density and type statistics of the points of interest within a certain range around each candidate location based on the point of interest data and the preliminary candidate location set, thereby obtaining the original data of the surrounding facilities; Step S253: weighting and scoring the original data of surrounding facilities according to the preset facility type weights, thereby obtaining surrounding facility scoring data; Step S254: Calculate the distance from each candidate location to the nearest substation and the surrounding power grid capacity according to the power grid distribution data, thereby obtaining the original data of the power grid capacity; and perform weighted calculation on the original data of the power grid capacity according to the preset power grid capacity weight, thereby obtaining the power grid capacity scoring data; Step S255: Perform a multi-factor comprehensive scoring based on the accessibility scoring data, the surrounding facilities scoring data, the power grid capacity scoring data, and the charging demand distribution map data, so as to obtain the candidate location comprehensive scoring data.

7. The method for planning charging piles by dynamic analysis according to claim 6, characterized in that: Step S255 includes the following steps: Step S2551: performing data alignment based on the preliminary candidate location set according to the accessibility score data, the surrounding facility score data, and the power grid capacity score data, and performing standardization processing to obtain a standardized score data set; Step S2552: performing weighted summation on the standardized scoring data set to obtain preliminary comprehensive scoring data; Step S2553: performing spatial autocorrelation analysis based on the spatial relationship between candidate positions according to the preliminary comprehensive scoring data, and performing spatial smoothing processing to obtain comprehensive scoring data; Step S2554: Perform demand-weighted adjustment on the comprehensive score data according to the charging demand distribution map data, so as to obtain comprehensive score data of the candidate locations.

8. The method for planning charging piles by dynamic analysis according to claim 7, characterized in that: Step S3 includes the following steps: Step S31: Acquire technical parameter data of different types of charging piles, including charging power, charging efficiency and interface type; Step S32: Obtain cost data of different types of charging piles, including equipment cost, installation cost, operation cost and maintenance cost; merge the technical parameter data and cost data into technical parameter and cost data; Step S33: establishing a charging pile performance-cost model according to the technical parameters and cost data, thereby obtaining charging pile performance-cost evaluation data; Step S34: setting an objective function of multi-objective optimization, including minimizing total cost, maximizing coverage, and maximizing service quality; determining the potential demand of each candidate location according to the location suitability score data and the charging demand distribution map data, thereby obtaining potential demand data; Step S35: Solve the objective function according to the charging pile performance-cost evaluation data and the potential demand data to obtain a non-dominated solution set; evaluate and screen the non-dominated solution set to obtain a preliminary configuration plan.

9. The method for planning charging piles by dynamic analysis according to claim 8, characterized in that: Step S4 includes the following steps: Step S41: Acquire power grid information data of the target area, including substation distribution, power line layout and load distribution; Step S42: constructing a power system capacity model according to the power grid information data, wherein constructing the power system capacity model specifically includes constructing a power grid topology model describing node and line information according to the power grid information data, establishing a line capacity limitation model according to the thermal capacity and stability limitation of the line, constructing a substation capacity limitation model based on the capacity constraint of the transformer and the switchgear, and integrating the power grid topology model, the line capacity limitation model and the substation capacity limitation model; Step S43: Analyze the grid carrying capacity of the preliminary configuration plan according to the power system capacity model to obtain grid load impact assessment data; Step S44: According to the grid load impact assessment data, the charging piles that do not meet the grid carrying capacity requirements are adjusted, and feasibility verification is performed to obtain an adjusted feasible configuration plan; Step S45: Simulate the operation status of the charging station in different time periods and scenarios according to the feasible configuration scheme, and perform performance evaluation based on the utilization rate, waiting time and grid load to obtain performance evaluation data.

10. The method for planning charging piles by dynamic analysis according to claim 9, characterized in that: Step S5 includes the following steps: Step S51: identifying the locations and time periods of charging stations that do not meet the preset performance targets based on the performance evaluation data, thereby obtaining performance target difference data; Step S52: taking the performance target as a constraint condition according to the preset performance target and the performance target difference data, and taking minimizing the adjustment cost as the objective function, thereby establishing a performance dynamic optimization model; Step S53: Design an optimization algorithm based on the adjustment of the number of charging piles, the replacement of the type, and the fine-tuning of the position according to the performance dynamic optimization model, and optimize the feasible configuration scheme using the optimization algorithm to obtain the final dynamic planning scheme.

Images