A charging station planning method and system based on vehicle charging demand

Abstract

The present invention belongs to the technical field of charging station planning. The present invention discloses a charging station planning method and system based on vehicle charging demands. By collecting and fusing multi-source data, a spatio-temporal charging demand distribution map is constructed. Seasonal change analysis and future demand prediction are carried out to obtain a dynamic charging demand prediction model. Combining with the urban road network structure and traffic flow data, the site layout is optimized under multiple constraints to form a preliminary charging station layout plan. The grid load capacity assessment and renewable energy access analysis are implemented to establish an energy collaborative supply guarantee system. The dynamic adjustment of charging prices during peak and valley periods and the appointment queuing mechanism are designed to form an intelligent scheduling control strategy. Finally, a diversified charging facility configuration plan is obtained through the evaluation of different charging power demands and vehicle type adaptability. The present invention overcomes the problems of inaccurate demand prediction, unreasonable layout, uneven resource allocation, etc. in traditional planning, and realizes the precise matching of charging resources and user demands.

Description

Technical Field

[0001] The present invention relates to the technical field of charging station planning. More specifically, the present invention relates to a charging station planning method and system based on vehicle charging demands. Background Art

[0002] With the increasing global environmental awareness and the in-depth promotion of carbon neutrality strategies in various countries, the electric vehicle industry is experiencing an unprecedented development wave. As a key support for the development of the electric vehicle industry, the planning and layout of charging infrastructure has become a core link in the transportation electrification strategy.

[0003] However, there are many problems to be solved in the existing charging station planning methods, which seriously restrict the healthy development of the electric vehicle industry. Traditional planning generally uses static population density or vehicle ownership for simple linear prediction, lacking refined analysis of users' actual travel trajectories and charging behaviors, resulting in a serious deviation of demand assessment from reality. In many areas, there is a shortage of charging piles, while in other areas, there are few users. Factors such as seasonal fluctuations, weather impacts, and policy changes are often ignored, making the planning lack foresight and adaptability and difficult to cope with the rapid changes in the electric vehicle market. Existing layouts are mostly based on administrative regions, ignoring urban functional zoning and traffic flow characteristics, resulting in unbalanced phenomena such as congestion during lunchtime in business districts and idle resources at night in residential areas. More seriously, the coordination between the construction of charging infrastructure and the power grid is insufficient, leading to frequent overload of regional power grids during peak periods and even causing power outage risks. The lack of an intelligent scheduling mechanism makes the allocation efficiency of charging resources low. Users often need to wait for a long time in popular areas, and the single price mechanism cannot effectively guide the dispersion of demand. At the same time, the standardization degree of existing facility configurations is low, and the charging demand differences of different vehicle models are not fully considered, resulting in problems such as insufficient allocation of ultra-fast charging resources and low utilization rate of slow charging piles, and the general poor charging experience of users. In actual operation, the phenomenon of data islands is serious, and there is no collaboration between stations, unable to provide users with a coherent service experience.

[0004] In view of this, the present invention proposes a charging station planning method and system based on vehicle charging demands to solve the above problems. Summary of the Invention

[0005] To overcome the above-mentioned defects of the prior art and to achieve the above object, the present invention provides the following technical solutions: In the first aspect, the present application provides a charging station planning method based on vehicle charging demands, including: Step 1: Collect and fuse multi-source data on vehicle driving trajectories and charging behaviors in a region to obtain a spatio-temporal charging demand distribution map;

[0006] Step 2: Conduct seasonal change analysis and future demand prediction based on the spatio-temporal charging demand distribution map to obtain a dynamic charging demand prediction model;

[0007] Step 3: Based on the dynamic charging demand prediction model, combined with the urban road network structure and traffic flow data, optimize the site layout under multiple constraints to obtain a preliminary layout plan for charging stations;

[0008] Step 4: Based on the preliminary layout plan for charging stations, conduct grid load capacity assessment and renewable energy access analysis to obtain an energy collaborative supply guarantee system;

[0009] Step 5: Based on the energy collaborative supply guarantee system, design dynamic adjustment of charging prices during peak and valley periods and a reservation queuing mechanism to obtain an intelligent scheduling control strategy;

[0010] Step 6: Based on the intelligent scheduling control strategy, conduct assessment of adaptability to different charging power requirements and vehicle models to obtain a diversified charging facility configuration plan.

[0011] On the other hand, the present application provides a charging station planning system based on vehicle charging demands, including:

[0012] A collection and processing module, configured to perform multi-source data collection and fusion processing on vehicle driving trajectories and charging behaviors within a region to obtain a spatio-temporal charging demand distribution map;

[0013] An analysis and demand prediction module, configured to perform seasonal change analysis and future demand prediction based on the spatio-temporal charging demand distribution map to obtain a dynamic charging demand prediction model;

[0014] A site layout optimization module, configured to optimize the site layout under multiple constraints based on the dynamic charging demand prediction model, combined with the urban road network structure and traffic flow data, to obtain a preliminary layout plan for charging stations;

[0015] An energy supply assessment module, configured to conduct grid load capacity assessment and renewable energy access analysis based on the preliminary layout plan for charging stations to obtain an energy collaborative supply guarantee system;

[0016] An intelligent scheduling control module, configured to design dynamic adjustment of charging prices during peak and valley periods and a reservation queuing mechanism based on the energy collaborative supply guarantee system to obtain an intelligent scheduling control strategy;

[0017] A facility configuration planning module, configured to conduct assessment of adaptability to different charging power requirements and vehicle models based on the intelligent scheduling control strategy to obtain a diversified charging facility configuration plan, and each module is connected by wired and / or wireless means.

[0018] The technical effects and advantages of the charging station planning method and system based on vehicle charging demands of the present invention:

[0019] By integrating multi-dimensional data and advanced algorithms, the present invention enhances the scientific nature and forward-looking nature of charging infrastructure planning, effectively solving the pain points in traditional planning methods such as inaccurate demand forecasting, unreasonable site layout, and unbalanced resource allocation. It realizes the precise matching of charging resources and user needs, significantly reducing the idle rate and congestion rate, and improving the utilization efficiency of charging facilities and the return on investment. Through the dynamic price mechanism and intelligent scheduling strategy, it not only balances the grid load and reduces the peak pressure, but also optimizes the energy distribution and reduces power waste. Special attention is paid to grid safety and the access of renewable energy, promoting the green transformation of the energy structure and reducing carbon emissions. For users, the charging experience is significantly improved, the waiting time is reduced, and the service quality is enhanced; for operators, the refined resource allocation and demand forecasting reduce the investment risk and improve the operational efficiency; for urban managers, the scientific and reasonable charging network layout promotes the electrification process of urban transportation and accelerates the construction of smart cities. In addition, the adaptability and scalability of the present invention enable it to continuously optimize with market changes and technological progress. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of a charging station planning method based on the charging demand of vehicles according to the present invention;

[0021] Figure 2 It is a schematic diagram of a charging station planning system based on the charging demand of vehicles according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0023] The embodiments of the present application provide a charging station planning method and system based on the charging demand of vehicles. The execution subjects of the charging station planning method and system based on the charging demand of vehicles include but are not limited to: vehicle networking platforms, traffic management systems, power grid management systems, data analysis platforms, planning decision support systems, etc. that are equipped with this system, which can be regarded as general computing nodes of the present application. The data processing platform includes but is not limited to at least one of: vehicle trajectory analysis systems, charging demand forecasting systems, and site optimization systems.

[0024] Please refer to Figure 1 , the present invention provides a charging station planning method based on the charging demand of vehicles, including the following steps:

[0025] Step 1: Collect and fuse multi-source data on vehicle driving trajectories and charging behaviors within the region to obtain a spatio-temporal charging demand distribution map;

[0026] Step 2: Conduct seasonal change analysis and future demand prediction based on the spatio-temporal charging demand distribution map to obtain a dynamic charging demand prediction model;

[0027] Step 3: Based on the dynamic charging demand prediction model, combined with the urban road network structure and traffic flow data, optimize the site layout under multiple constraints to obtain a preliminary layout plan for charging stations;

[0028] Step 4: Conduct grid load capacity assessment and renewable energy access analysis based on the preliminary layout plan for charging stations to obtain an energy collaborative supply guarantee system;

[0029] Step 5: Based on the energy collaborative supply guarantee system, design a dynamic adjustment of charging prices during peak and valley periods and a reservation queuing mechanism to obtain an intelligent scheduling control strategy;

[0030] Step 6: Based on the intelligent scheduling control strategy, conduct an assessment of the adaptability of different charging power requirements and vehicle models to obtain a diversified charging facility configuration plan.

[0031] The present invention accurately captures the regional charging demand through multi-source data collection and fusion processing, constructs a spatio-temporal charging demand distribution map to provide a data basis for subsequent analysis, conducts seasonal change analysis and future demand prediction based on the spatio-temporal charging demand distribution map to make the planning forward-looking, the dynamic charging demand prediction model can adapt to demand changes at different time scales, optimizes the site layout under multiple constraints by combining the urban road network structure and traffic flow data to improve the scientific nature of site selection, the preliminary layout plan for charging stations considers various actual factors to ensure feasibility, solves the energy supply problem through grid load capacity assessment and renewable energy access analysis, the energy collaborative supply guarantee system enhances the reliability and sustainability of the charging network, the dynamic adjustment of charging prices during peak and valley periods and the design of the reservation queuing mechanism optimize the resource utilization efficiency, the intelligent scheduling control strategy improves the user experience and system operation efficiency, and the assessment of the adaptability of different charging power requirements and vehicle models ensures the universality of the service, and the diversified charging facility configuration plan meets the diverse needs of different users.

[0032] In the embodiment of the present invention, the detailed implementation steps of Step 1 include:

[0033] Obtain vehicle GPS trajectory data from the vehicle networking platform to get a vehicle movement trajectory set, and conduct data cleaning and outlier detection on the vehicle movement trajectory set to obtain effective trajectory data;

[0034] Obtain historical charging records from the charging pile operator management system to get a charging behavior dataset, and extract time features and spatial features from the charging behavior dataset to obtain a charging mode feature library;

[0035] Obtain road traffic flow monitoring data from the traffic management department to get traffic flow distribution information, and overlay the traffic flow distribution information with the urban functional area division layer to obtain a regional activity hotspot map;

[0036] Conduct spatio-temporal correlation analysis on the valid trajectory data, charging mode feature library, and regional activity hotspot map to obtain charging demand association rules, and construct a hierarchical clustering model based on the charging demand association rules to obtain a charging demand clustering result;

[0037] Visualize and map the charging demand clustering result on a geographic information system to obtain a spatio-temporal charging demand distribution map.

[0038] In this embodiment, first, the driving trajectory data of electric vehicles in the area is collected. By establishing a data interface with the vehicle networking platform, information such as the GPS location, timestamp, speed, and direction of the vehicle is obtained, forming a vehicle movement trajectory set containing a large number of trajectory points. Data cleaning is performed on the obtained vehicle movement trajectory set, including removing null values, correcting GPS drift, repairing data breakpoints, and unifying the time format. Statistical methods or machine learning algorithms (such as isolation forest, LOF algorithm, etc.) are used to detect and process outliers, such as trajectory points that deviate significantly from the normal route. Finally, high-quality valid trajectory data is obtained. Then, historical charging records are collected from the management systems of each charging pile operator, including information such as charging start / end time, charging location, charging power, charging duration, user ID, etc., forming a charging behavior data set. Time feature analysis is performed on the charging behavior data set, such as charging peak periods, average charging duration, periodic patterns, etc. Time series analysis methods are used to extract time patterns. Spatial feature analysis is performed on the charging behavior data set, such as hot spot area distribution, regional preference, spatial aggregation, etc. Spatial statistical methods are used to extract spatial patterns. The charging mode feature library is formed by integrating spatio-temporal features. At the same time, road traffic flow monitoring data is obtained from the traffic management department, including information such as road traffic volume, congestion conditions, traffic events, etc., which is collected through multi-source channels such as traffic monitoring cameras, traffic volume detectors, and traffic App user data, generating traffic flow distribution information. The traffic flow distribution information is subjected to GIS overlay analysis with the urban functional area division layer (such as commercial areas, residential areas, industrial areas, office areas, etc.) to identify the areas and time periods with intensive human activities, forming a regional activity hot spot map. Then, spatio-temporal data mining technology is used to perform correlation analysis on the valid trajectory data, charging mode feature library, and regional activity hot spot map, discovering the association rules between vehicle activity patterns and charging behaviors, such as "when the vehicle density in the commercial area increases by 30% from 3 to 5 pm on weekdays, the charging demand around it rises by 50%", etc., forming charging demand association rules. Based on the charging demand association rules, hierarchical clustering algorithms (such as DBSCAN, HDBSCAN, etc.) are used to perform clustering analysis on the charging demand, clustering similar charging demand patterns into one category according to time similarity and spatial proximity, forming a charging demand clustering result. Finally, the charging demand clustering result is imported into the Geographic Information System (GIS), and through visualization methods such as heat maps, contour maps, and 3D topographic maps, the charging demand distribution in different regions and different time periods is displayed, forming an intuitive spatio-temporal charging demand distribution map. This map clearly shows the charging demand change pattern of "when, where, and to what extent", providing a data basis for subsequent charging station planning.

[0039] In the embodiment of the present invention, the detailed implementation steps of step 2 include:

[0040] Slice the spatio-temporal charging demand distribution map according to the time dimension to obtain demand sequences with different time granularities, including intra-day, intra-week, monthly, and seasonal sequences;

[0041] Perform time series decomposition on the demand sequences with different time granularities to obtain trend terms, seasonal terms, and random terms, and identify the growth pattern of the trend terms to obtain the long-term development trend function;

[0042] Extract the periodic characteristics of the seasonal terms to obtain a seasonal change pattern library, and conduct correlation analysis in combination with meteorological data to obtain a climate factor influence model;

[0043] Collect data on the market penetration rate of electric vehicles and policy changes to obtain a set of market growth driving factors, and combine the set of market growth driving factors with the long-term development trend function to obtain a market development prediction curve;

[0044] Based on the long-term development trend function, seasonal change pattern library, climate factor influence model, and market development prediction curve, construct a multi-level time neural network model to obtain a dynamic charging demand prediction model.

[0045] In this embodiment, first, the data in the spatio-temporal charging demand distribution map is sliced according to different time granularities to extract the intra-day change sequences (such as hourly changes), analyze features such as morning and evening rush hours and working hours, extract the intra-week change sequences (such as differences between weekdays and weekends), analyze the demand pattern differences between weekdays and rest days, extract the monthly change sequences, analyze the demand change rules at the beginning, middle, and end of the month, extract the seasonal sequences, analyze the demand change patterns in the four seasons of spring, summer, autumn, and winter, form demand sequences with different time granularities. Then, time series decomposition techniques (such as STL decomposition, X-12-ARIMA, etc.) are used to decompose the demand sequences with different time granularities into three components: long-term trend term, periodic seasonal term, and random fluctuation term. By decomposition, the pattern becomes clearer. For the decomposed trend term, growth pattern recognition algorithms (such as linear regression, exponential smoothing, polynomial regression, etc.) are used to identify the long-term growth pattern of the charging demand, such as linear growth, exponential growth, or S-shaped growth, etc., and a mathematical function expression is fitted to form a long-term development trend function. For the decomposed seasonal term, periodic analysis methods (such as Fourier analysis, wavelet analysis, etc.) are used to extract the change patterns with fixed periods, such as daily cycle patterns, weekly cycle patterns, seasonal cycle patterns, etc., to form a seasonal change pattern library. Historical meteorological data such as temperature, precipitation, humidity, and extreme weather are obtained from the meteorological department, the correlation between meteorological factors and charging demand is analyzed, and an impact model of climate conditions on charging demand is established, such as the amplitude of increase or decrease in charging demand during extreme weather, to form a climate factor impact model. At the same time, electric vehicle market data is collected, including electric vehicle sales volume, growth rate, regional distribution, etc., and policy change data is collected, including subsidy policies, purchase restriction policies, environmental protection regulations, etc., and the driving effects of these factors on charging demand are analyzed to form a set of market growth driving factors. The set of market growth driving factors is integrated and analyzed with the long-term development trend function, and scenario analysis or statistical modeling methods are used to predict the future electric vehicle market size and growth trajectory to form a market development prediction curve. Finally, the long-term development trend function, seasonal change pattern library, climate factor impact model, and market development prediction curve are integrated, and deep learning techniques are used to construct a multi-level time neural network model, such as LSTM (Long Short-Term Memory Network), GRU (Gated Recurrent Unit), Transformer, etc. The model has a hierarchical structure with multiple time scales and can capture both short-term fluctuations and long-term trends simultaneously. Through training and verification with historical data, a dynamic charging demand prediction model is formed. This model can dynamically predict the future charging demand changes according to multiple factors such as time, location, climate, and market, providing a scientific basis for charging station planning.

[0046] In the embodiment of the present invention, the detailed implementation steps of step 3 include:

[0047] Obtain the road network topology structure and traffic flow data within the area, obtain a weighted road network graph, and divide traffic analysis areas on the weighted road network graph to obtain a candidate area set;

[0048] Conduct a preliminary screening of the candidate area set in combination with the land use plan and the power grid distribution situation to obtain a feasible site set, and conduct an accessibility analysis on the feasible site set to obtain a service coverage map;

[0049] Based on the dynamic charging demand prediction model, calculate the potential service demand of each candidate site to obtain a demand weight matrix, and construct a multi-objective optimization function, including the objectives of demand satisfaction, coverage balance, and construction cost minimization;

[0050] Apply the improved particle swarm optimization algorithm to solve the multi-objective optimization function to obtain a Pareto optimal solution set, and conduct a comprehensive evaluation of the Pareto optimal solution set through the fuzzy analytic hierarchy process to obtain the optimal site configuration plan;

[0051] Conduct service quality verification and sensitivity analysis on the optimal site configuration plan to obtain a preliminary layout plan for charging stations.

[0052] In this embodiment, first, obtain the road network topology structure data within the region from the traffic management department and the urban planning department, including information such as road grades, connection relationships, lengths, and traffic capacities. Collect traffic flow data, including vehicle flow, congestion conditions, average driving speeds, etc. at different times. Combine the road network topology structure with the traffic flow data to construct a weighted road network graph, where road connections represent the topology structure and weights represent the traffic costs (such as time cost, distance cost). Divide the region on the weighted road network graph according to traffic characteristics and administrative divisions, such as dividing by traffic cells, functional blocks, or grid patterns, to form a set of candidate regions as possible charging station site selection regions. Then, combine with the urban land use planning data to screen out regions that meet the land use nature requirements, such as commercial land, public facility land, parking lot land, etc. Collect power grid distribution data, including substation locations, distribution line distributions, power capacities, etc., and screen out regions with better power grid access conditions. Mark the regions that meet both land and power grid conditions to form a set of feasible site sets. Use network analysis methods (such as the shortest path algorithm, service area analysis, etc.) to perform reachability analysis on the set of feasible site sets, calculate the coverage range that can be covered within a specific time (such as 5 minutes, 10 minutes) starting from each potential site, form a service coverage range graph, and display the service radiation range and covered population of each site. Use a dynamic charging demand prediction model, combined with the service coverage range graph, to calculate the potential charging demand at the locations of each candidate site, consider various indicators such as daily average and peak demand, assign different demand weights to different locations to form a demand weight matrix, construct a multi-objective optimization function, and comprehensively consider three key objectives: demand satisfaction (maximizing the charging demand coverage rate), coverage balance (evenly distributing services to avoid service blind spots), and construction cost minimization (considering land costs, power grid access costs, etc.). Reflect the importance of different objectives through the setting of objective function weights. Use an improved particle swarm optimization (PSO) algorithm to solve the multi-objective optimization problem, introduce improvement mechanisms such as crowding distance and elite strategy to improve the algorithm performance, and find the best site combination through multiple iterative optimizations to obtain a Pareto optimal solution set composed of a set of non-dominated solutions. Each solution represents a possible site layout plan. Conduct a comprehensive evaluation of each solution in the Pareto optimal solution set through the fuzzy analytic hierarchy process, set up an evaluation index system and determine the index weights, calculate the comprehensive scores of each plan through fuzzy comprehensive evaluation, and select the plan with the highest score as the optimal site configuration plan. Finally, verify the service quality of the optimal site configuration plan, use queuing theory models or simulation methods to evaluate the service capabilities and waiting times at different times, conduct sensitivity analysis, evaluate the impact of changes in key parameters (such as electric vehicle penetration rate, charging demand, etc.) on the layout plan, and make necessary adjustments to the plan according to the verification and analysis results. Finally, form a preliminary layout plan for the charging stations, which not only meets the charging demand but also has economic feasibility and implementation flexibility.

[0053] In the embodiments of the present invention, the detailed implementation steps of step 4 include:

[0054] Obtain the topological structure of the regional distribution network and the substation capacity data to obtain the power grid basic information library, and simulate the access situation of the charging load according to the preliminary layout plan of the charging stations to obtain the power grid power flow distribution model;

[0055] Conduct N-1 security verification and voltage stability analysis on the power grid power flow distribution model to obtain the power grid bearing capacity assessment report, and identify the weak links and potential risk points of the power grid based on the power grid bearing capacity assessment report to obtain the power grid enhancement requirement list;

[0056] Investigate the distribution and generation characteristics of renewable energy in the region according to the power grid enhancement requirement list to obtain the renewable energy resource map, and evaluate the types and scales of renewable energy suitable for access to each charging station based on the renewable energy resource map to obtain the green energy configuration plan;

[0057] Design the configuration parameters and energy management strategy of the energy storage system based on the green energy configuration plan and the power grid power flow distribution model to obtain the peak shaving and valley filling solution, and integrate the peak shaving and valley filling solution with the green energy configuration plan to obtain the hybrid energy system architecture;

[0058] Build a collaborative optimization scheduling model of the distribution network with charging station groups based on the hybrid energy system architecture to obtain the multi-time scale energy scheduling strategy, and verify the system reliability of the multi-time scale energy scheduling strategy through digital twin simulation to obtain the energy collaborative supply guarantee system.

[0059] In this embodiment, first, obtain the topological structure data of the regional distribution network from the power company, including network structure information such as substation locations, line routes, and circuit breaker locations, and collect substation capacity data, including main transformer capacity, load rate, standby capacity, etc. Integrate them to form a power grid basic information database. According to the preliminary layout plan of the charging stations, calculate the expected load characteristics of each charging station, including peak power, load curve, power factor, etc. Use power system analysis software (such as PSASP, PowerFactory, etc.) to connect the charging load to the power grid for simulation, simulate the operation state of the power grid at different times and different load levels, generate a power grid power flow distribution model, conduct N-1 security verification on the power grid power flow distribution model, that is, simulate the operation state of the system when any one key component (such as a transformer, line) fails, evaluate the safety margin of the system, conduct voltage stability analysis, calculate indicators such as voltage deviation and voltage stability margin of each node. Based on the comprehensive results of security verification and stability analysis, form a power grid carrying capacity assessment report. This report comprehensively evaluates the ability of the power grid to carry the load of charging stations. Analyze and identify the weak links in the power grid according to the assessment report, such as substations with insufficient capacity, overloaded lines, areas with low voltage, etc., evaluate potential risk points, such as peak load impact, harmonic pollution, impact load, etc., and form a power grid enhancement demand list, including equipment that needs to be upgraded and transformed, newly added power facilities, etc. Investigate the distribution of renewable energy in the region, such as areas rich in solar energy resources, areas rich in wind energy resources, geothermal resource distribution, etc., analyze the power generation characteristics of different renewable energy sources, such as output characteristics, seasonal changes, stability, etc., and integrate them to form a renewable energy resource map. Based on the renewable energy resource map, combined with the geographical location and land use conditions of each charging station, evaluate the suitable types (such as photovoltaic, wind, biomass energy, etc.) and scales of renewable energy to be connected. Considering technical feasibility, economy, and environmental impact, form a green energy configuration plan for each charging station. According to the green energy configuration plan and the power grid power flow distribution model, design the configuration parameters of the energy storage system, including energy storage capacity, power characteristics, response speed, etc., formulate an energy management strategy, such as peak-valley price arbitrage, demand-side response, load transfer, etc., and form a peak shaving and valley filling solution, which can suppress the charging load fluctuation and improve the operation efficiency of the power grid. Integrate the peak shaving and valley filling solution with the green energy configuration plan to design the overall architecture of the hybrid energy system, including energy conversion equipment, control system, communication interface, etc., and form a hybrid energy system architecture. Regard each charging station as an overall network, construct a coordinated optimization scheduling model for the distribution network containing a charging station group, consider factors such as power grid constraints, renewable energy output characteristics, and dynamic changes in charging demand, design multi-time scale energy scheduling strategies, including day-ahead scheduling, hourly scheduling, and real-time scheduling, to achieve cross-time scale energy optimization allocation. Use digital twin technology to construct a virtual simulation environment to comprehensively simulate and verify the multi-time scale energy scheduling strategies.Test the reliability and stability of the system under different scenarios (such as peak periods, off-peak periods, extreme weather, etc.). Through repeated testing and optimization, an energy collaborative supply guarantee system is finally formed. This system can ensure the safe, economic, and environmental protection of the energy supply for the charging station network, providing an energy foundation for the next step of intelligent scheduling control.

[0060] In the embodiment of the present invention, the detailed implementation steps of step 5 include:

[0061] Analyze the regional power grid load curve and electricity price policy to obtain the basic electricity price structure, and combine the basic electricity price structure with the charging demand time distribution characteristics to determine the peak-valley period division standard, and obtain the initial time-of-use electricity price plan;

[0062] Based on the initial time-of-use electricity price plan, establish a quantitative model for the price sensitivity of charging behavior to obtain the demand price elasticity coefficient, and use the demand price elasticity coefficient to design a dynamic price adjustment algorithm to obtain a real-time pricing mechanism;

[0063] Based on the real-time pricing mechanism and the energy collaborative supply guarantee system, construct a charging station resource pre-allocation model to obtain the predicted service capacity value, and design a priority-based reservation rule system according to the predicted service capacity value to obtain a reservation management framework;

[0064] Combine the reservation management framework to develop a personalized charging plan recommendation system based on vehicle characteristics and user preferences to obtain an intelligent recommendation engine, and design a dynamic queuing scheduling algorithm under congestion conditions based on the intelligent recommendation engine to obtain a queue optimization model;

[0065] According to the multi-time scale energy scheduling strategy of the energy collaborative supply guarantee system, integrate the real-time pricing mechanism, the reservation management framework, the intelligent recommendation engine, and the queue optimization model to obtain an intelligent scheduling control strategy.

[0066] In this embodiment, first, historical load curve data of the regional power grid are collected to analyze the load change rules and characteristics, such as intra-day fluctuations, seasonal variations, etc., and the local electricity price policies are understood, including existing time-of-use electricity prices, stepped electricity prices, demand response policies, etc. The basic electricity price structure is integrated. Combining the basic electricity price structure with the charging demand time distribution characteristics obtained in Step 2, the division criteria for peak, flat, and valley periods of charging services are determined, such as peak periods (10:00 - 15:00, 18:00 - 21:00), flat periods (7:00 - 10:00, 15:00 - 18:00, 21:00 - 23:00), and valley periods (23:00 - 7:00 the next day). Different electricity price levels are set according to the characteristics of each period to form an initial time-of-use electricity price plan. Through the analysis of historical charging data, a quantitative model of the sensitivity of charging behavior to price changes is established, and econometric methods (such as regression analysis, discrete choice models, etc.) are used to estimate the price elasticity of demand coefficients for different user groups and different periods. For example, the elasticity coefficient during peak periods is -0.8 (when the price increases by 10%, the demand decreases by 8%), and the elasticity coefficient during valley periods is -1.2 (price drops by 10%, demand increases by 12%), based on the demand price elasticity coefficient, design a dynamic price adjustment algorithm, which can automatically adjust the price according to the real-time grid load status, renewable energy output and charging demand forecast. The algorithm adopts reinforcement learning or model predictive control method to form a real-time pricing mechanism. Using the real-time pricing mechanism and the energy supply situation in the energy collaborative supply guarantee system, a charging station resource pre-allocation model is constructed to predict the maximum service capacity in different time periods. Considering the hardware constraints such as the number and power of charging piles and the energy constraints such as grid capacity and renewable energy output, a service capacity forecast value is formed. According to the service capacity forecast value, a priority-based reservation rule system is designed, such as long-distance user priority, low-power vehicle priority, VIP member priority and other rules. The reservation time window, cancellation policy, credit mechanism and other details are designed to form a reservation management framework. The characteristics of different electric vehicles (such as battery capacity, charging rate, cruising range, etc.) and user charging preferences (such as charging time, cost sensitivity, time flexibility, etc.) are analyzed to develop individual The personalized charging solution recommendation system can recommend the optimal charging time, location and charging amount for users. It uses machine learning algorithms (such as collaborative filtering, content recommendation, etc.) to achieve personalized recommendations and form an intelligent recommendation engine. In view of the congestion status of charging stations, a dynamic queuing scheduling algorithm is designed to optimize the charging sequence and resource allocation by considering factors such as the urgency of vehicle charging demand, waiting time, and charging duration. The algorithm can use heuristic methods or dynamic programming methods to maximize system throughput and user satisfaction and form a queue optimization model. Finally, according to the multi-time scale energy scheduling strategy in the energy collaborative supply guarantee system, the real-time pricing mechanism, reservation management framework, intelligent recommendation engine and queue optimization model are systematically integrated, and a unified control interface and information sharing mechanism are developed to ensure that each subsystem works together to achieve full-process intelligent scheduling from energy supply to user service, and form an intelligent scheduling control strategy. This strategy can achieve efficient utilization of charging resources, continuous optimization of user experience, and effective management of power grid load, providing intelligent operation support for the charging station network. .

[0067] In the embodiment of the present invention, the detailed implementation steps of step 6 include:

[0068] Collect the parameters of the main electric vehicle models circulating in the market to obtain a model database, and classify the model database by charging interface standards and charging power requirements according to the intelligent scheduling control strategy to obtain the model grouping results;

[0069] Based on the vehicle type grouping results, the target user group characteristics of each charging station are analyzed to obtain user portraits. The proportion distribution of different charging power requirements is predicted based on the user portraits to obtain the power demand structure.

[0070] Combine the power demand structure with the physical space constraints of the charging station and the power capacity limit to obtain the site resource constraint conditions, and construct an optimization model for facility configuration based on the site resource constraint conditions to obtain a preliminary configuration plan;

[0071] Based on the queuing model analysis of the preliminary configuration plan and the intelligent scheduling control strategy, considering the reasonable ratio of fast charging to slow charging and the reserved space for future technology upgrades, obtain the facility structure optimization strategy, and adjust the preliminary configuration plan based on the facility structure optimization strategy to obtain a flexible expansion plan;

[0072] Use the service quality evaluation indicators in the intelligent scheduling control strategy to comprehensively evaluate the preliminary configuration plan and the flexible expansion plan, balance the investment benefits and user experience, and obtain a diversified charging facility configuration plan.

[0073] In this embodiment, first, technical parameters of mainstream electric vehicle models in the market are collected, including different types such as passenger vehicles, commercial vehicles, and buses. Key parameters such as battery capacity, maximum charging power, charging interface type, and charging curve characteristics of each model are recorded to form a structured vehicle model database. According to the user data and charging records in the intelligent scheduling control strategy, the vehicle model database is analyzed, classified according to charging interface standards (such as national standards, European standards, Tesla, etc.) and charging power requirements (such as ultra-fast charging, fast charging, slow charging), and the main vehicle model groups and their characteristics are identified to form the vehicle model clustering results. Based on the vehicle model clustering results and the location characteristics of charging stations (such as commercial areas, residential areas, highways, etc.), the target user groups that each charging station may serve are analyzed. Considering factors such as vehicle type, usage scenario, and charging habits, user portraits are constructed, such as "Charging stations in commercial areas mainly serve office commuting vehicles, are frequently used on weekdays, and stay for 4 - 8 hours". According to the user portraits, the demand ratios of different charging stations for different power charging facilities are analyzed, such as "Highway service areas need 80% ultra-fast charging + 20% fast charging", "Residential areas need 30% fast charging + 70% slow charging", etc., to form the power demand structure. The physical space conditions of each charging station are investigated, including available area, parking space layout, electrical equipment room space, etc., and the power capacity limitations are evaluated, including transformer capacity, line carrying capacity, distribution equipment limitations, etc. The power demand structure is combined with physical space constraints and power capacity limitations for analysis to clarify the resource constraint conditions of each station. Based on the station resource constraint conditions, an optimal facility configuration model is constructed. Considering factors such as construction cost, operation efficiency, and user satisfaction, linear programming or integer programming methods are used to determine the optimal quantity ratio of various charging devices, such as the number of 7kW slow charging piles, 60kW fast charging piles, 120kW ultra-fast charging piles, etc., to form a preliminary configuration plan. Combining the queuing model in the intelligent scheduling control strategy, the possible congestion situation and service efficiency of the charging station are analyzed. Based on the analysis results, the ratio of fast charging to slow charging is adjusted to ensure service capacity during peak periods, and future technology upgrade space is reserved, such as reserving interfaces and power capacity for ultra-high power charging of 350kW and above, to form an optimized facility structure strategy. According to the optimized facility structure strategy, the preliminary configuration plan is adjusted to increase modular and scalable design, enabling the system to be flexibly upgraded according to demand changes. A phased construction plan is designed, and a multi-stage construction plan is formulated according to the predicted demand growth curve to form a flexible expansion plan. Using the service quality evaluation indicators in the intelligent scheduling control strategy, such as average waiting time, charging pile utilization rate, user satisfaction, etc., the preliminary configuration plan and the flexible expansion plan are comprehensively evaluated. Through multi-scenario simulations, the performance of different plans is verified, and the investment benefits (such as construction cost, payback period) and user experience (such as convenience, reliability) are balanced to select the optimal plan, forming a diversified charging facility configuration plan that can meet the diverse charging needs of different vehicle models and different users, while ensuring the economy of investment and the scalability of the system.

[0074] In the embodiments of the present invention, a comprehensive evaluation is performed on the Pareto optimal solution set through the fuzzy analytic hierarchy process to obtain an optimal site configuration plan, including:

[0075] Establish a charging station evaluation index system, including multi-dimensional evaluation indexes such as demand coverage rate, service radius balance, traffic accessibility, grid access convenience, and land resource utilization efficiency, obtain an evaluation standard framework, and determine the ideal values and acceptable thresholds of each evaluation index to obtain an index reference interval;

[0076] Based on the index reference interval, compare the importance of each evaluation index, express the comparison results using linguistic variables, obtain a fuzzy semantic evaluation set, and convert the fuzzy semantic evaluation set into triangular fuzzy numbers to construct a fuzzy judgment matrix to obtain an expert judgment structure;

[0077] Adopt the geometric mean method to synthesize multiple expert judgment structures to obtain a comprehensive judgment matrix, and use the fuzzy extension analysis method to calculate the fuzzy weight vector of each evaluation index to obtain the quantization result of index importance;

[0078] Conduct a consistency test on the comprehensive judgment matrix, calculate the consistency ratio to obtain the judgment reliability evaluation result. If the consistency requirement is not met, return to readjust the expert judgment structure until the consistency requirement is met to obtain an effective weight vector;

[0079] Perform fuzzy scoring on each solution plan in the Pareto optimal solution set for each evaluation index to obtain a plan scoring matrix, and combine the effective weight vector to calculate the comprehensive evaluation value of each plan to obtain a plan ranking result;

[0080] Select the plan with the highest comprehensive evaluation value as the basic plan, and conduct a feasibility verification of actual implementation, including on-site investigation and stakeholder consultation, to obtain implementation constraint conditions, and fine-tune the basic plan according to the implementation constraint conditions to finally form an optimal site configuration plan.

[0081] In this embodiment, first, a multi-dimensional index system for evaluating charging stations is established, including multiple dimensions such as demand coverage rate (the degree to which a charging station can meet the regional charging demand), service radius balance (the degree of balanced distribution of the service scope of a charging station), traffic accessibility (the convenience of the charging station's traffic and the road connectivity), grid access convenience (the ease and cost of accessing the grid), and land resource utilization efficiency (the service capacity per unit area of land), etc., to form an evaluation standard framework. Through expert discussions and case analyses, the ideal values (such as demand coverage rate ≥ 95%) and acceptable thresholds (such as demand coverage rate ≥ 80%) of each evaluation index are determined to form an index reference interval, providing a standard basis for subsequent evaluations. Invite experts in the fields of charging station planning, grid construction, urban transportation, etc. to conduct pairwise importance comparisons of each evaluation index based on the index reference interval, and use linguistic variables (such as "extremely important", "very important", "slightly important", "equally important", etc.) to express the comparison results, forming a fuzzy semantic evaluation set. Convert the fuzzy semantic evaluation set into triangular fuzzy numbers, such as "extremely important" is converted to (7, 9, 9), "very important" is converted to (5, 7, 9), etc., construct a judgment matrix containing triangular fuzzy numbers to form an expert judgment structure. When multiple experts participate in the evaluation, use the geometric mean method to synthesize the judgment structures of multiple experts, calculate the geometric mean of each judgment element to form a comprehensive judgment matrix, use the Fuzzy Extent Analysis method to process the comprehensive judgment matrix, calculate the fuzzy weight vector of each evaluation index, and obtain the clear weight value of each index through defuzzification (such as the centroid method) of the fuzzy weight vector to form the quantitative result of index importance. Conduct a consistency test on the comprehensive judgment matrix, calculate the consistency index (CI) and consistency ratio (CR), and judge the degree of consistency of expert evaluations. If CR < 0.If it is 1, it is considered that the judgment has acceptable consistency. Otherwise, it is necessary to return and ask the experts to readjust the judgment until the consistency requirement is met, forming an effective weight vector. Fuzzy scoring is performed on each site layout plan in the Pareto optimal solution set for each evaluation index. The scoring result can be represented by triangular fuzzy numbers to reflect the uncertainty of the evaluation, forming a plan scoring matrix. Combine the plan scoring matrix with the effective weight vector, calculate the comprehensive evaluation value of each plan. The fuzzy weighted average method or other fuzzy multi-attribute decision-making methods can be used. Sort all the plans according to the comprehensive evaluation value to obtain the plan sorting result. Select the plan with the highest comprehensive evaluation value as the basic plan, organize a professional team to conduct on-site surveys to verify the feasibility of the plan in the actual environment, consult stakeholders such as local governments, community residents, and power grid companies, collect their opinions and suggestions, identify the possible constraint conditions in the implementation process, such as the difficulty of land acquisition, the acceptance of residents, the construction period, etc. Make necessary fine-tuning to the basic plan according to the implementation constraint conditions, such as adjusting the positions of some sites, adjusting the construction time sequence, adding supporting facilities, etc., while retaining the core advantages of the original plan and improving the implementation feasibility, and finally form an optimal site configuration plan, which not only meets the requirements of multi-dimensional evaluation indicators but also has practical operability.

[0082] In the embodiment of the present invention, a charging station resource pre-allocation model is constructed based on the real-time pricing mechanism and the energy collaborative supply guarantee system to obtain a service capacity prediction value, including:

[0083] Obtain historical charging demand data and price sensitivity indicators to obtain a price-demand response relationship matrix, and combine the dynamic price adjustment parameters of the real-time pricing mechanism to establish a demand transfer prediction model to obtain a load distribution prediction map;

[0084] Extract the grid capacity constraint and the renewable energy supply fluctuation characteristics from the energy collaborative supply guarantee system to obtain the energy supply boundary conditions, and construct a multi-period supply capacity evaluation model based on the energy supply boundary conditions to obtain the time-varying capacity upper limit;

[0085] Perform matching analysis on the load distribution prediction map and the time-varying capacity upper limit to obtain the supply-demand balance constraint conditions, and develop a dynamic resource allocation algorithm based on the supply-demand balance constraint conditions to obtain a preliminary resource allocation plan;

[0086] Based on the preliminary resource allocation plan, establish a dual-objective optimization function considering user satisfaction and system efficiency to obtain the resource allocation optimization goal, and use the rolling horizon optimization method to solve the dual-objective optimization function to obtain an optimized resource allocation strategy;

[0087] Design an uncertainty processing mechanism based on stochastic processes in combination with an optimized resource allocation strategy, conduct scenario simulations on the demand prediction errors in the load distribution prediction map and the renewable energy fluctuations in the energy supply boundary conditions, obtain risk assessment results, and adjust the resource reservation ratio in the optimized resource allocation strategy according to the risk assessment results to form a resilient service capacity prediction value.

[0088] In this embodiment, historical charging demand data is first collected and analyzed, including charging demand at different price levels, charging time period distribution, etc. Price sensitivity indicators such as price elasticity coefficient and cross-price elasticity are calculated, and a response relationship matrix between price and demand is constructed to show the impact of different price changes on charging demand at different time periods. Combining with the dynamic price adjustment parameters in the real-time pricing mechanism (such as benchmark price, adjustment range, adjustment frequency, etc.), a demand transfer prediction model is established. This model can predict the transfer effect of charging demand in time and space caused by price changes, such as "a 20% increase in peak-hour price will cause 30% of the demand to transfer to off-peak periods". Time series analysis and machine learning methods (such as regression models, neural networks, etc.) are used to construct the prediction model, and a load distribution prediction map is formed to show the spatio-temporal distribution prediction of charging demand under different price strategies. Grid-related constraints such as transformer capacity limits, line transmission limits, and voltage stability boundaries are extracted from the energy collaborative supply guarantee system, and the supply characteristics of renewable energy such as the daily curve of photovoltaic power generation, the fluctuation characteristics of wind power, and seasonal changes are extracted. These are integrated to form energy supply boundary conditions, and the maximum available energy supply at different times and locations is clarified. Based on the energy supply boundary conditions, a multi-period supply capacity evaluation model is constructed, considering factors such as grid load level, renewable energy output prediction, and energy storage system status, to calculate the maximum power supply capacity for each time period, forming a time-varying capacity upper limit, and showing the maximum power supply capacity curve for each time period within 24 hours. The load distribution prediction map is matched and analyzed with the time-varying capacity upper limit to identify potential supply-demand imbalance points, such as time periods or regions where demand exceeds supply capacity, and supply-demand balance constraint conditions are formed, including hard constraints that must be met (such as not exceeding the maximum grid capacity) and soft constraints that should be met as much as possible (such as giving priority to high-value users). Based on the supply-demand balance constraint conditions, a dynamic resource allocation algorithm is developed. This algorithm can calculate the amount of charging resources to be pre-allocated for each time period according to the predicted demand and available resources. Considering time continuity and resource finiteness, linear programming or network flow algorithms are used for resource allocation optimization to form a preliminary resource allocation plan. Based on the preliminary resource allocation plan, a two-objective optimization function is established. On the one hand, it maximizes user satisfaction (such as reducing waiting time and meeting preferred time periods), and on the other hand, it maximizes system efficiency (such as improving equipment utilization rate and balancing grid load). Weight coefficients are set for the two objectives to reflect their relative importance, forming a resource allocation optimization goal. The rolling horizon optimization method is used to solve the two-objective optimization function. The optimization time period is divided into multiple rolling horizons (such as one horizon every 3 hours). First, the recent horizon is optimized, and then it is gradually advanced backward. Each time it advances, the prediction data and boundary conditions are updated. This method can balance the optimization effect and computational complexity to form an optimized resource allocation strategy. Considering the uncertainties in actual operation, a processing mechanism based on stochastic processes is designed, and methods such as Monte Carlo simulation are used to generate multiple scenarios of demand prediction errors and renewable energy fluctuations.Evaluate the robustness of the resource allocation strategy under uncertain conditions to form a risk assessment result. According to the risk assessment result, adjust and optimize the resource reservation ratio in the resource allocation strategy. For example, increase the standby resources by 5-10% during high-uncertainty periods to ensure that the system has sufficient response capabilities. Finally, form a resilient service capacity prediction value that not only considers the expected demand but also includes a buffer space for coping with uncertainties, providing a reliable basis for the resource reservation and scheduling of charging stations.

[0089] The above describes the charging station planning method based on vehicle charging demand in the embodiments of the present application. Next, the charging station planning system based on vehicle charging demand in the embodiments of the present application will be described. Please refer to Figure 2 One embodiment of the charging station planning system based on vehicle charging demand in the embodiments of the present application includes:

[0090] A collection and processing module for collecting and fusing multi-source data on vehicle driving trajectories and charging behaviors in the area to obtain a spatio-temporal charging demand distribution map;

[0091] An analysis and demand prediction module for performing seasonal change analysis and future demand prediction based on the spatio-temporal charging demand distribution map to obtain a dynamic charging demand prediction model;

[0092] A site layout optimization module for optimizing the site layout under multiple constraints based on the dynamic charging demand prediction model in combination with the urban road network structure and traffic flow data to obtain a preliminary charging station layout plan;

[0093] An energy supply assessment module for evaluating the grid load capacity and analyzing the access of renewable energy based on the preliminary charging station layout plan to obtain an energy collaborative supply guarantee system;

[0094] An intelligent scheduling control module for dynamically adjusting the charging price during peak and valley periods and designing a reservation and queuing mechanism based on the energy collaborative supply guarantee system to obtain an intelligent scheduling control strategy;

[0095] A facility configuration planning module for evaluating the adaptability of different charging power requirements and vehicle types based on the intelligent scheduling control strategy to obtain a diversified charging facility configuration plan. Each module is connected by wired and / or wireless means to achieve data transmission between modules.

[0096] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

[0097] It should be noted that in this text, the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, article or device including the said element.

[0098] In the description of the present invention, it should be understood that terms such as "first", "second", etc. are only used for descriptive distinction and cannot be construed as indicating or implying relative importance.

[0099] In the description of the present invention, unless otherwise specified, the meaning of "a plurality" is two or more.

[0100] In the description of the present invention, the meaning of "several" is one or more, and the meaning of "a large number" is two or more.

[0101] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. mean that the specific features, structures, materials or characteristics described in connection with the said embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0102] For the formulas in this specification, the dimensional units are removed and only the numerical values are calculated. The formulas are obtained by collecting a large amount of data and performing software simulation to get a formula closest to the actual situation. The preset parameters and threshold values in the formulas are set by those skilled in the art according to the actual situation.

[0103] Although embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the claims and their equivalents.

Claims

1. A charging station planning method based on the charging demand of automobiles, characterized in that, Including: Step 1: Collect and fuse multi-source data on vehicle driving trajectories and charging behaviors within a region to obtain a spatio-temporal charging demand distribution map; including: Obtain vehicle GPS trajectory data from the vehicle networking platform to get a vehicle movement trajectory set, and perform data cleaning and outlier detection on the vehicle movement trajectory set to obtain valid trajectory data; Obtain historical charging records from the charging pile operator management system to get a charging behavior data set, and extract time features and spatial features from the charging behavior data set to obtain a charging mode feature library; Obtain road traffic flow monitoring data from the traffic management department to get traffic flow distribution information, and overlay the traffic flow distribution information with the urban functional area division layer to obtain a regional activity hotspot map; Perform spatio-temporal correlation analysis on the valid trajectory data, the charging mode feature library, and the regional activity hotspot map to obtain charging demand association rules, and construct a hierarchical clustering model based on the charging demand association rules to obtain a charging demand clustering result; Perform visual mapping of the charging demand clustering result on a geographic information system to obtain a spatio-temporal charging demand distribution map; Step 2: Conduct seasonal change analysis and future demand prediction based on the spatio-temporal charging demand distribution map to obtain a dynamic charging demand prediction model, including: Slice the spatio-temporal charging demand distribution map according to the time dimension to obtain demand sequences with different time granularities, including intra-day, intra-week, monthly, and seasonal sequences; Decompose the demand sequences with different time granularities into trend terms, seasonal terms, and random terms, and identify the growth pattern of the trend terms to obtain a long-term development trend function; Extract periodic characteristics of the seasonal terms to obtain a seasonal change pattern library, and conduct correlation analysis in combination with meteorological data to obtain a climate factor influence model; Collect data on the penetration rate of the electric vehicle market and policy changes to get a set of market growth driving factors, and combine the set of market growth driving factors with the long-term development trend function to obtain a market development prediction curve; Based on the long-term development trend function, the seasonal change pattern library, the climate factor influence model, and the market development prediction curve, construct a multi-level time neural network model to obtain a dynamic charging demand prediction model; Step 3: Based on the dynamic charging demand prediction model, combine the urban road network structure and traffic flow data to optimize the site layout under multiple constraints to obtain a preliminary charging station layout plan; Step 4: Conduct grid load capacity assessment and renewable energy access analysis based on the preliminary charging station layout plan to obtain an energy collaborative supply guarantee system; Step 5: Based on the energy collaborative supply guarantee system, design a dynamic adjustment of charging prices during peak and valley periods and a reservation queuing mechanism to obtain an intelligent scheduling control strategy; Step 6: Based on the intelligent scheduling control strategy, conduct an assessment of the adaptability of different charging power demands and vehicle models to obtain a diversified charging facility configuration plan.

2. The charging station planning method based on vehicle charging requirements according to claim 1, wherein Based on the dynamic charging demand prediction model, combined with the urban road network structure and traffic flow data, optimize the site layout under multiple constraints to obtain a preliminary layout plan for charging stations, including: Obtain the topological structure of the road network and traffic flow data in the region to get a weighted road network graph, and divide traffic analysis areas on the weighted road network graph to obtain a set of candidate areas; Conduct a preliminary screening of the set of candidate areas in combination with land use planning and grid distribution conditions to obtain a set of feasible site addresses, and conduct an accessibility analysis of the set of feasible site addresses to obtain a service coverage map; Based on the dynamic charging demand prediction model, calculate the potential service demand of each candidate site to obtain a demand weight matrix, and construct a multi-objective optimization function, including objectives of demand satisfaction, coverage balance, and minimization of construction costs; Apply an improved particle swarm optimization algorithm to solve the multi-objective optimization function to obtain a Pareto optimal solution set, and conduct a comprehensive evaluation of the Pareto optimal solution set through the fuzzy analytic hierarchy process to obtain an optimal site configuration plan; Conduct service quality verification and sensitivity analysis on the optimal site configuration plan to obtain a preliminary layout plan for charging stations.

3. The charging station planning method based on vehicle charging demand according to claim 2, characterized in that, Based on the preliminary layout plan for charging stations, conduct grid load capacity assessment and renewable energy access analysis to obtain an energy collaborative supply guarantee system, including: Obtain the topological structure of the regional distribution network and substation capacity data to get a grid basic information database, and simulate the charging load access situation according to the preliminary layout plan for charging stations to obtain a grid power flow distribution model; Conduct N-1 security verification and voltage stability analysis on the grid power flow distribution model to obtain a grid carrying capacity assessment report, and identify grid weak links and potential risk points based on the grid carrying capacity assessment report to obtain a grid reinforcement demand list; Investigate the distribution and generation characteristics of renewable energy in the region according to the grid reinforcement demand list to obtain a renewable energy resource map, and evaluate the types and scales of renewable energy suitable for access to each charging site based on the renewable energy resource map to obtain a green energy configuration plan; Based on the green energy configuration plan and the grid power flow distribution model, design the configuration parameters and energy management strategies of the energy storage system to obtain a peak shaving and valley filling solution, and integrate the peak shaving and valley filling solution with the green energy configuration plan to obtain a hybrid energy system architecture; Based on the hybrid energy system architecture, construct a coordinated optimization scheduling model for the distribution network with charging station clusters to obtain a multi-time scale energy scheduling strategy, and verify the system reliability of the multi-time scale energy scheduling strategy through digital twin simulation to obtain an energy collaborative supply guarantee system.

4. The charging station planning method based on automotive charging requirements according to claim 3, wherein Based on the energy collaborative supply guarantee system, conduct dynamic adjustment of charging prices during peak and valley periods and design a reservation queuing mechanism to obtain an intelligent scheduling control strategy, including: Analyze the regional grid load curve and electricity price policy to obtain a basic electricity price structure, and combine the basic electricity price structure with the time distribution characteristics of charging demand to determine the division criteria for peak, flat, and valley periods to obtain an initial time-of-use electricity price plan; Based on the initial time-of-use electricity price plan, establish a quantitative model for the price sensitivity of charging behavior, obtain the demand price elasticity coefficient, and use the demand price elasticity coefficient to design a dynamic price adjustment algorithm to obtain a real-time pricing mechanism; Based on the real-time pricing mechanism and the energy collaborative supply guarantee system, construct a charging station resource pre-allocation model, obtain the service capacity prediction value, and design a priority-based reservation rule system according to the service capacity prediction value to obtain a reservation management framework; Combine the reservation management framework to develop a personalized charging plan recommendation system based on vehicle characteristics and user preferences, obtain an intelligent recommendation engine, and design a dynamic queuing scheduling algorithm under congested conditions based on the intelligent recommendation engine to obtain a queue optimization model; According to the multi-time scale energy scheduling strategy of the energy collaborative supply guarantee system, integrate the real-time pricing mechanism, the reservation management framework, the intelligent recommendation engine, and the queue optimization model into a system to obtain an intelligent scheduling control strategy.

5. The charging station planning method based on vehicle charging requirements according to claim 4, wherein Perform different charging power demand and vehicle type adaptability evaluations based on the intelligent scheduling control strategy to obtain a diversified charging facility configuration plan, including: Collect the parameters of the main electric vehicle models circulating in the market to obtain a vehicle model database, and classify the charging interface standards and charging power requirements of the vehicle model database according to the intelligent scheduling control strategy to obtain the vehicle model clustering results; Analyze the target user group characteristics of each charging station based on the vehicle model clustering results to obtain user portraits, and predict the proportional distribution of different charging power requirements according to the user portraits to obtain the power demand structure; Combine the power demand structure with the physical space constraints and power capacity limits of the charging station to obtain the site resource constraint conditions, and construct a facility configuration optimization model based on the site resource constraint conditions to obtain a preliminary configuration plan; According to the analysis of the queuing model of the preliminary configuration plan and the intelligent scheduling control strategy, consider the reasonable ratio of fast charging to slow charging and the reserved space for future technology upgrades to obtain a facility structure optimization strategy, and adjust the preliminary configuration plan based on the facility structure optimization strategy to obtain a flexible expansion plan; Use the service quality evaluation index in the intelligent scheduling control strategy to comprehensively evaluate the preliminary configuration plan and the flexible expansion plan, balance the investment benefit and user experience, and obtain a diversified charging facility configuration plan.

6. The charging station planning method based on vehicle charging demand according to claim 2, wherein Comprehensively evaluate the Pareto optimal solution set through the fuzzy analytic hierarchy process to obtain an optimal site configuration plan, including: Establish a charging station evaluation index system, including multi-dimensional evaluation indexes such as demand coverage rate, service radius balance, traffic accessibility, grid access convenience, and land resource utilization efficiency, obtain an evaluation standard framework, and determine the ideal values and acceptable thresholds of each evaluation index to obtain an index reference interval; Compare the importance of each evaluation index based on the index reference interval, express the comparison results using linguistic variables to obtain a fuzzy semantic evaluation set, and convert the fuzzy semantic evaluation set into triangular fuzzy numbers to construct a fuzzy judgment matrix to obtain an expert judgment structure; The geometric mean method is used to synthesize multiple expert judgment structures to obtain a comprehensive judgment matrix, and the fuzzy extended analysis method is used to calculate the fuzzy weight vector of each evaluation index to obtain the quantitative result of the index importance; The consistency of the comprehensive judgment matrix is tested, the consistency ratio is calculated, and the judgment reliability evaluation result is obtained. If the consistency requirement is not met, the expert judgment structure is returned for readjustment until the consistency requirement is met, and an effective weight vector is obtained; Each solution in the Pareto optimal solution set is subject to fuzzy scoring on each evaluation index to obtain a solution scoring matrix, and the comprehensive evaluation value of each solution is calculated in combination with the effective weight vector to obtain the solution ranking result; The solution with the highest comprehensive evaluation value is selected as the basic solution, and the practical implementation feasibility is verified, including on-site investigation and stakeholder consultation, to obtain the implementation constraint conditions. The basic solution is fine-tuned according to the implementation constraint conditions to finally form the optimal site configuration solution.

7. The charging station planning method based on vehicle charging demand according to claim 4, wherein Based on the real-time pricing mechanism and the energy collaborative supply guarantee system, a charging station resource pre-allocation model is constructed to obtain the service capacity prediction value, including: Historical charging demand data and price sensitivity indicators are obtained to obtain a price-demand response relationship matrix, and in combination with the dynamic price adjustment parameters of the real-time pricing mechanism, a demand transfer prediction model is established to obtain a load distribution prediction map; The grid capacity constraint and the renewable energy supply fluctuation characteristics are extracted from the energy collaborative supply guarantee system to obtain the energy supply boundary conditions, and based on the energy supply boundary conditions, a multi-period supply capacity evaluation model is constructed to obtain the time-varying capacity upper limit; The load distribution prediction map is matched and analyzed with the time-varying capacity upper limit to obtain the supply-demand balance constraint conditions, and based on the supply-demand balance constraint conditions, a dynamic resource allocation algorithm is developed to obtain a preliminary resource allocation plan; Based on the preliminary resource allocation plan, a dual-objective optimization function considering user satisfaction and system efficiency is established to obtain the resource allocation optimization goal, and the rolling time domain optimization method is used to solve the dual-objective optimization function to obtain the optimized resource allocation strategy; Combined with the optimized resource allocation strategy, an uncertainty processing mechanism based on a stochastic process is designed to perform scenario simulation on the demand prediction error in the load distribution prediction map and the renewable energy fluctuation in the energy supply boundary conditions to obtain the risk assessment result, and the resource reservation ratio in the optimized resource allocation strategy is adjusted according to the risk assessment result to form a resilient service capacity prediction value.

8. A charging station planning system based on vehicle charging requirements, which is used to implement the charging station planning method based on vehicle charging requirements according to any one of claims 1 to 7, characterized in that Including: A collection and processing module for multi-source data collection and fusion processing of vehicle driving trajectories and charging behaviors in the region to obtain a spatio-temporal charging demand distribution map; An analysis and demand prediction module for seasonal change analysis and future demand prediction based on the spatio-temporal charging demand distribution map to obtain a dynamic charging demand prediction model; A site layout optimization module for optimizing the site layout under multiple constraints based on the dynamic charging demand prediction model in combination with the urban road network structure and traffic flow data to obtain a preliminary charging station layout plan; An energy supply evaluation module, which is used to evaluate the grid load capacity and analyze the access of renewable energy based on the preliminary layout plan of the charging station, so as to obtain an energy collaborative supply guarantee system; An intelligent scheduling and control module, which is used to dynamically adjust the charging price during peak and valley periods and design a reservation and queuing mechanism based on the energy collaborative supply guarantee system, so as to obtain an intelligent scheduling and control strategy; A facility configuration planning module, which is used to evaluate the charging power requirements and vehicle type adaptability based on the intelligent scheduling and control strategy, so as to obtain a diversified charging facility configuration plan. Each module is connected by wired and / or wireless means.

Images