Urban charging pile distribution reachability analysis method and system

Abstract

The invention relates to the technical field of intelligent traffic and urban planning, and provides an urban charging pile distribution accessibility analysis method and system, and the method comprises the steps: collecting the related data and historical charging demand data of an existing charging pile; respectively constructing a potential position identification model and a demand prediction model, and obtaining a charging pile potential position identification result and a charging pile demand prediction result; a reachability analysis model is constructed, and a reachability analysis result is obtained through the reachability analysis model in combination with the charging pile potential position identification result and the charging pile demand prediction result; a cost-benefit analysis model is constructed based on the cost-benefit analysis framework, and a plurality of charging pile layout schemes are obtained in combination with the accessibility analysis result; and establishing a corresponding traffic simulation model to obtain an optimal charging pile layout scheme, thereby solving the problems that the existing method lacks comprehensive consideration capability, is difficult to accurately predict the future charging demand, and cannot effectively evaluate the cost effectiveness and accessibility of different charging pile layout schemes.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent transportation and urban planning, and in particular to a method and system for analyzing the distribution accessibility of urban charging piles. Background Art

[0002] With the popularization of electric vehicles and the acceleration of urbanization, the distribution and accessibility of urban charging piles have become important factors restricting the development of electric vehicles. Reasonable layout of charging piles can not only meet the growing demand for charging, but also improve the convenience and economy of electric vehicle use and promote the green transformation of urban transportation. However, the current distribution of urban charging piles is often blind, lacking scientific and systematic planning and evaluation methods, resulting in uneven utilization of charging piles and mismatch between charging demand and supply.

[0003] Traditional charging pile layout methods mainly rely on empirical judgment and simple data analysis, and lack the ability to comprehensively consider multiple factors such as road network, traffic flow, electric vehicle distribution, population density, and historical charging demand. This method often fails to accurately predict future charging demand, and cannot effectively evaluate the cost-effectiveness and accessibility of different charging pile layout options. Summary of the invention

[0004] The purpose of the present invention is to provide a photovoltaic power generation energy storage grid-connected optimization system and method, aiming to solve the problem that the existing charging pile layout method mainly relies on experience judgment and simple data analysis, lacks comprehensive consideration ability, is difficult to accurately predict future charging demand, and cannot effectively evaluate the cost-effectiveness and accessibility of different charging pile layout schemes.

[0005] The embodiments of the present invention are implemented by the following technical solutions:

[0006] A method for analyzing the distribution accessibility of charging piles in a city, comprising the following steps:

[0007] Collect relevant data of existing charging piles and historical charging demand data, perform corresponding data preprocessing on the relevant data of existing charging piles and historical charging demand data, and obtain charging pile relevant data sets and historical charging demand data sets respectively;

[0008] Based on the charging pile related data set and the historical charging demand data set, a potential location identification model and a demand prediction model are constructed respectively, and the charging pile potential location identification results and charging pile demand prediction results are obtained;

[0009] Construct an accessibility analysis model, combine the potential location identification results of charging piles and the demand prediction results of charging piles, and obtain the accessibility analysis results through the accessibility analysis model;

[0010] A cost-benefit analysis framework is set up based on the relevant data of existing charging piles and historical charging demand data. A cost-benefit analysis model is constructed based on the cost-benefit analysis framework. Combined with the accessibility analysis results, several charging pile layout plans are obtained.

[0011] Based on several charging pile layout schemes, a corresponding traffic simulation model is established. Each charging pile layout scheme is simulated through the traffic simulation model and the corresponding simulation results are obtained. Each simulation result is compared to obtain the optimal charging pile layout scheme.

[0012] Optionally, the relevant data of the existing charging piles include road network data, traffic flow data, electric vehicle distribution data and population density data; the historical charging demand data include user historical density data, electric vehicle historical density data and charging pile historical utilization data.

[0013] Optionally, based on the charging pile related data set, a potential location recognition model is constructed, and the specific process of obtaining the charging pile potential location recognition result is as follows:

[0014] Through the charging pile related data set, the starting point, end point, length and speed limit of each road in the road data are obtained;

[0015] A potential location recognition model is constructed based on graph theory. The starting point and end point of each road are added as nodes to the potential location recognition model, and the length and speed limit of each road are attached as edge attributes.

[0016] Set the length and speed limit constraints, calculate the edge weights based on the constraints, and update the edge attributes;

[0017] The importance of each node is evaluated by degree centrality, the average degree centrality of each node is calculated, and nodes with degree centrality greater than the average are selected as potential charging pile locations.

[0018] Optionally, based on the historical charging demand data set, a demand forecasting model is constructed to obtain the charging pile demand forecasting result in the following specific process:

[0019] Extract the basic characteristics and time characteristics of charging demand through the historical charging demand data set, and calculate the corresponding derived characteristics based on the basic characteristics and time characteristics of charging demand;

[0020] The corresponding demand feature dataset is constructed through the basic features, time features and derived features of charging demand, and the demand feature dataset is divided into training set, validation set and test set according to the time sequence;

[0021] Based on gradient boosting tree and geographically weighted regression, an initial demand forecasting model is constructed, basic parameters of the initial demand forecasting model are set, the initial demand forecasting model is trained through the training set, the initial demand forecasting model is evaluated through the validation set, model parameters of the initial demand forecasting model are tuned according to the evaluation results and the optimal demand forecasting model is obtained, the optimal demand forecasting model is tested through the test set and the demand forecasting model and the corresponding charging pile demand forecasting results are obtained.

[0022] Optionally, the demand feature data set includes:

[0023] Feature set, as input data of demand forecasting model, where each sample contains basic features, time features and derived features of charging demand;

[0024] The target variable set, which serves as the output data of the demand forecasting model, is a set of target variables, in which each element is associated with a corresponding sample in the feature set.

[0025] Optionally, the specific process of constructing the accessibility analysis model, combining the charging pile potential location identification result and the charging pile demand prediction result, and obtaining the accessibility analysis result through the accessibility analysis model is as follows:

[0026] Based on GIS, a reachability analysis model is constructed through Di jkstra algorithm and A* search algorithm;

[0027] Combining the potential location identification results of charging piles and the demand prediction results of charging piles, the weights of the required number of charging piles, user density and travel time are initialized;

[0028] Based on the heuristic site selection algorithm, the accessibility analysis results are obtained through the accessibility analysis model.

[0029] Optionally, the specific process of setting a cost-benefit analysis framework according to relevant data of existing charging piles and historical charging demand data, constructing a cost-benefit analysis model based on the cost-benefit analysis framework, and obtaining several charging pile layout solutions in combination with the accessibility analysis results is as follows:

[0030] Set up a cost-benefit analysis framework based on the relevant data of existing charging piles and historical charging demand data;

[0031] Based on the cost-benefit analysis framework, a cost-benefit analysis model is constructed. The total cost and total benefit are calculated through the cost-benefit analysis model. According to the time series data of total cost and total benefit, the net present value and internal rate of return of the project are calculated, and the cost-benefit analysis results are obtained.

[0032] By combining the results of cost-benefit analysis with those of accessibility analysis, several charging pile layout schemes are obtained.

[0033] Optionally, the specific process of establishing a corresponding traffic simulation model based on several charging pile layout schemes, simulating each charging pile layout scheme through the traffic simulation model and obtaining corresponding simulation results, and comparing each simulation result to obtain the optimal charging pile layout scheme is as follows:

[0034] Based on several charging pile layout schemes, set the corresponding traffic model parameters and charging pile model parameters, dynamically simulate the charging pile layout through traffic simulation software and establish the corresponding traffic simulation model;

[0035] Dynamically simulate each charging pile layout plan through the traffic simulation model, collect the corresponding data in each dynamic simulation process, evaluate the performance of each charging pile layout, adjust the location, number or type of charging piles according to the evaluation results, and optimize the charging pile layout plan;

[0036] Conduct sensitivity analysis on the charging pile model parameters in the optimized charging pile layout scheme to observe the impact of different parameter changes on the optimization effect of the charging pile layout scheme;

[0037] Compare the performance of each optimized charging pile layout solution and select the best charging pile layout solution;

[0038] Implement the selected charging pile layout plan, establish a monitoring system, collect relevant data of the charging piles and provide feedback, and iteratively optimize the system through the collected charging pile-related data.

[0039] The present invention also provides a system for implementing the urban charging pile distribution accessibility analysis method, comprising:

[0040] Data collection and preprocessing module, used to collect relevant data of existing charging piles and historical charging demand data, and perform data preprocessing;

[0041] Model building module, used to build potential location identification model, demand forecasting model and accessibility analysis model based on preprocessed data;

[0042] Cost-benefit analysis module, which is used to build a cost-benefit analysis model and generate multiple charging pile layout plans based on existing data and accessibility analysis results;

[0043] Traffic simulation and optimization module, which is used to establish a traffic simulation model, simulate the traffic conditions under different charging pile layout schemes, and optimize the optimal layout scheme;

[0044] The result display and decision support module is used to display the analysis results in a visual form and provide intuitive data support and decision suggestions for decision makers.

[0045] Optionally, the data acquisition and preprocessing module includes:

[0046] A data collection submodule is used to obtain road network data, traffic flow data, electric vehicle distribution data, population density data, user historical density data, electric vehicle historical density data and charging pile historical utilization rate data from data sources;

[0047] The data preprocessing submodule is used to clean, integrate and format the collected data to generate charging pile related data sets and historical charging demand data sets;

[0048] The model building module includes:

[0049] Potential location identification submodule: used to identify potential charging pile locations using graph theory and degree centrality evaluation methods;

[0050] The demand forecasting submodule is used to build and optimize the demand forecasting model based on gradient boosting tree and geographically weighted regression to predict the demand for charging piles;

[0051] The accessibility analysis submodule is used to build an accessibility analysis model based on GIS and heuristic site selection algorithm to evaluate the accessibility of charging pile layout;

[0052] The cost-benefit analysis module includes:

[0053] The cost-benefit analysis framework setting submodule is used to set the basic framework of cost-benefit analysis according to data and requirements;

[0054] The cost-benefit calculation submodule is used to calculate the total cost and total benefit and conduct net present value and internal rate of return analysis;

[0055] The layout scheme generation submodule is used to generate multiple charging pile layout schemes by combining the accessibility analysis results and the cost-benefit analysis results;

[0056] The traffic simulation and optimization module includes:

[0057] Traffic simulation model building submodule, used to build traffic simulation model based on traffic flow theory and simulation software;

[0058] The simulation operation and result analysis submodule is used to simulate the traffic conditions under different charging pile layout schemes and analyze the simulation results;

[0059] The optimization submodule is used to compare the advantages and disadvantages of different layout schemes according to the simulation results and select the optimal charging pile layout scheme;

[0060] The result display and decision support module includes:

[0061] The result visualization submodule is used to display the charging pile layout plan, accessibility analysis results and cost-benefit analysis results in the form of charts, maps, etc.

[0062] The decision support submodule is used to provide decision makers with specific suggestions for optimizing the layout of charging piles based on the analysis results.

[0063] The technical solution of the embodiment of the present invention has at least the following advantages and beneficial effects:

[0064] By collecting and analyzing the relevant data of existing charging piles and historical charging demand data, potential charging pile locations can be accurately identified and future charging demand can be predicted, which helps to avoid blind layout of charging piles, ensure that charging piles can cover real demand hotspots, and improve the utilization efficiency of charging piles and the rationality of the overall layout;

[0065] Constructing an accessibility analysis model, combined with the potential location identification results of charging piles and the charging demand prediction results, can scientifically evaluate the accessibility of different charging pile layout schemes, and by optimizing the layout, reduce the user's charging travel time and cost, improve the user experience, and further promote the popularization of electric vehicles;

[0066] Setting up a cost-benefit analysis framework and building a corresponding analysis model can comprehensively consider the total cost and total benefit of the charging pile layout. By calculating the net present value and internal rate of return of the project, it can provide decision makers with an economically feasible layout plan, which helps to maximize economic benefits while ensuring social benefits;

[0067] By comprehensively using a variety of technical means such as data processing, model building, and simulation, a scientific and systematic analysis method is provided for the layout of charging piles, which can not only improve the scientificity and accuracy of decision-making, but also reduce the subjectivity and uncertainty in the decision-making process;

[0068] By optimizing the layout of charging piles and improving the convenience and economy of electric vehicle charging, it will help promote the popularization of electric vehicles. As a clean energy transportation tool, the popularization of electric vehicles will effectively reduce carbon emissions from urban transportation and promote the green and sustainable development of urban transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 Schematic diagram of the process of analyzing the distribution accessibility of charging piles in a city in Embodiment 1 of the present invention;

[0070] Figure 2 This is a structural diagram of the urban charging pile distribution accessibility analysis system in Example 2 of the present invention. DETAILED DESCRIPTION

[0071] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0072] Example 1

[0073] Reference Figure 1 , a method for analyzing the distribution accessibility of urban charging piles, comprising the following steps:

[0074] Step 1: Collect relevant data of existing charging piles and historical charging demand data, perform corresponding data preprocessing on the relevant data of existing charging piles and historical charging demand data, and obtain charging pile relevant data sets and historical charging demand data sets respectively;

[0075] In this embodiment, the relevant data of existing charging piles include road network data, traffic flow data, electric vehicle distribution data and population density data; the historical charging demand data include user historical density data, electric vehicle historical density data and charging pile historical utilization rate data;

[0076] By investigating the needs and behavior patterns of electric vehicle users and collecting road network data, traffic flow data, electric vehicle distribution data, and population density data, we can obtain the geographic coordinates of possible locations for charging stations at road intersections, important buildings, commercial areas, residential areas, etc.

[0077] Road network data includes road type (such as highways, main roads, secondary roads, branch roads, sidewalks, etc.), length, width, directionality, connectivity (i.e. which roads are connected to each other), intersection information, bridge and tunnel locations, etc.; it can be collected and maintained through a GIS (Geographic Information System) system.

[0078] Traffic flow data refers to information such as the number, type, speed, and direction of travel of vehicles passing through a specific road or intersection within a certain period of time; it can be collected through a variety of methods such as road monitoring cameras, ground sensing coils, GPS tracking systems, and mobile phone positioning data.

[0079] Electric vehicle distribution data includes information such as the number and type of electric vehicles, the location of charging stations, the geographical distribution of users, and charging behavior; data that can be collected through government subsidy programs, automobile manufacturers, charging station operators, and market research and consumer surveys.

[0080] Clean and format the collected relevant data to ensure data quality, and perform spatial processing on the data, for example, convert non-spatial data into GIS-compatible formats.

[0081] Step 2: Based on the charging pile related data set and the historical charging demand data set, a potential location recognition model and a demand prediction model are constructed respectively, and the charging pile potential location recognition results and the charging pile demand prediction results are obtained;

[0082] In this embodiment, the starting point, end point, length and speed limit of each road in the road data are obtained through the charging pile related data set;

[0083] A potential location recognition model is constructed based on graph theory. The starting point and end point of each road are added as nodes to the potential location recognition model, and the length and speed limit of each road are attached as edge attributes.

[0084] Set the length and speed limit constraints, calculate the edge weights based on the constraints, and update the edge attributes;

[0085] The importance of each node is evaluated by degree centrality, the average degree centrality of each node is calculated, and nodes with degree centrality greater than the average are selected as potential charging pile locations.

[0086] First, define a list containing city road information, where each road is a dictionary, and each dictionary contains the starting point, end point, length, and speed limit; use the NetworkX library to create an undirected graph, which is used to represent the city road network, where nodes represent intersections, important landmarks, or potential charging pile locations, and edges represent road connections; by traversing each road in the list, add the road as an edge to the graph. Each edge is attached with additional attributes, including the length and speed limit of the road. Traverse each edge in the graph and calculate the weight of the edge based on the length and speed limit of the road. The weight of the edge is calculated by dividing the length of the road by the speed limit. The calculated edge weight attribute is added to each edge, so that in subsequent analysis, it is convenient to evaluate the importance of the edge based on the weight of the edge. Use the corresponding function in NetworkX to calculate the degree centrality of each node in the graph. In order to determine which nodes have relatively high degree centrality, calculate the average degree centrality of all nodes. Traverse the degree centrality dictionary and select nodes with degree centrality higher than the average as potential charging pile locations. These nodes are usually located in the center of the network or on important traffic nodes, so they are ideal locations for installing charging piles. Verify that the model is built correctly by printing the node and edge information of the graph. This helps confirm that all roads have been correctly added to the graph and the weights of the edges have been set correctly. Use the drawing function of NetworkX in combination with Matplotlib to visualize the constructed latent location recognition model. Through visualization, you can intuitively see the layout of the roads, the location of the nodes, and the connection relationship between them. Finally, use the corresponding function of Matplotlib to display the graph.

[0087] In this embodiment, the basic characteristics and time characteristics of charging demand are extracted from the historical charging demand data set, and corresponding derived characteristics are calculated based on the basic characteristics and time characteristics of charging demand;

[0088] The corresponding demand feature data set is constructed through the basic characteristics, time characteristics and derived characteristics of charging demand, and the demand feature data set is divided into training set, validation set and test set according to the time series; when dividing the data set, it is necessary to ensure that the training set, validation set and test set are continuous in time to avoid data leakage, that is, the data in the test set should be after the training set and validation set.

[0089] Based on gradient boosting tree and geographically weighted regression, an initial demand forecasting model is constructed, basic parameters of the initial demand forecasting model are set, the initial demand forecasting model is trained through the training set, the initial demand forecasting model is evaluated through the validation set, model parameters of the initial demand forecasting model are tuned according to the evaluation results and the optimal demand forecasting model is obtained, the optimal demand forecasting model is tested through the test set and the demand forecasting model and the corresponding charging pile demand forecasting results are obtained.

[0090] In this embodiment, the feature set is used as the input data of the demand forecasting model, wherein each sample includes basic features, time features and derived features of charging demand; the basic features of charging demand include features such as user density, electric vehicle density and charging pile utilization rate; the time features are extracted from the timestamp, including time features such as hours, dates, weeks and months, which are used to capture the periodic changes of time series data; the derived features can be calculated by the basic features and time features of charging demand, and the derived features include working days, whether it is a holiday, daily charging demand average, weekly charging demand average, monthly charging demand average and other derived features.

[0091] The target variable set, which serves as the output data of the demand forecasting model, is a set of target variables, in which each element is associated with a corresponding sample in the feature set.

[0092] Step 3: Construct an accessibility analysis model, combine the potential location identification results of charging piles and the demand prediction results of charging piles, and obtain the accessibility analysis results through the accessibility analysis model;

[0093] In this embodiment, based on GIS, a reachability analysis model is constructed by using the Di jkstra algorithm and the A* search algorithm;

[0094] Combining the potential location identification results of charging piles and the demand prediction results of charging piles, the weights of the required number of charging piles, user density and travel time are initialized;

[0095] Based on the heuristic site selection algorithm, the accessibility analysis results are obtained through the accessibility analysis model.

[0096] Prepare a weight list of nodes based on user density and an adjacency list representation of the graph for calculating travel time; determine the number k of charging piles that the algorithm needs to select and the weights of user density and travel time; use the Di jkstra algorithm to calculate the shortest travel time from each node to other nodes; evaluate the weight of each node based on user density and accessibility; based on the weight of the node, select the first k nodes as the location of the charging pile.

[0097] Step 4: Set up a cost-benefit analysis framework based on the relevant data of existing charging piles and historical charging demand data, build a cost-benefit analysis model based on the cost-benefit analysis framework, and combine the accessibility analysis results to obtain several charging pile layout plans;

[0098] In this embodiment, a cost-benefit analysis framework is set based on relevant data of existing charging piles and historical charging demand data;

[0099] The cost-benefit analysis framework includes:

[0100] Determine the analysis objectives and scope, specify the evaluation objectives of return on investment, social benefits, and environmental benefits, and specify the geographical boundaries of the analysis, the time span of the project, and the scale of the project;

[0101] Identify cost elements, including direct cost identification, indirect cost identification and cost allocation; direct cost identification includes specific costs for equipment purchase and installation and construction, indirect cost identification includes hidden costs such as land leasing, purchase, long-term operation and maintenance, and management, and cost allocation is the annual allocation of long-term costs for annual cost assessment;

[0102] Identify the benefit factors, including economic benefit assessment, environmental benefit assessment and social benefit assessment; among them, economic benefit assessment is to calculate the expected income from charging service charges, environmental benefit assessment is to estimate the reduced carbon emissions and try to monetize them, and social benefit assessment is to analyze the positive impact of improved traffic efficiency and the popularization of electric vehicles;

[0103] Cost estimation methods include market price research, bill of quantities and historical data comparison; market price research is to collect current market prices of equipment and construction, bill of quantities is to list all engineering tasks and related costs, and historical data comparison does not refer to the construction costs of similar projects and make cost forecasts;

[0104] Benefit quantification methods include environmental benefit quantification and time value quantification; environmental benefit quantification is to calculate environmental benefits based on the reduced carbon emissions multiplied by the price in the carbon trading market, and time value quantification is to estimate the time value saved by reducing traffic congestion;

[0105] Consideration of time value includes selecting a discount rate and calculating present value; selecting a discount rate is to determine the discount rate based on factors such as market interest rates and project risks, and calculating present value is to discount the expected future cash flows and calculate their present value;

[0106] Sensitivity analysis, including determining the variable range and simulation analysis; determining the variable range is to set the range of changes in variables such as charging pile utilization, electricity prices and policy changes, and simulation analysis is to observe the impact of key cover variables on the economic benefits of the project;

[0107] Risk assessment, including identification of risk factors, risk probability and impact assessment, and risk response strategies; among them, identification of risk factors is to list all potential risks, such as technical failures and changes in market demand, etc., risk probability and impact assessment is to assess the possibility of risk factors and their impact on the project, and risk response strategies are to formulate corresponding risk mitigation measures and alternative plans;

[0108] Calculation of financial indicators, including NPV calculation, IRR determination, payback period and benefit index; NPV calculation is to calculate the sum of the present values ​​of all cash flows during the project life cycle, IRR determination is to determine the discount rate that makes NPV equal to zero through trial and error or financial software, payback period is to calculate the time required for the investment cost to be recovered through project benefits, and benefit index is to calculate the ratio of project benefits to investment costs and evaluate the project's benefit potential.

[0109] Based on the cost-benefit analysis framework, a cost-benefit analysis model is constructed. The total cost and total benefit are calculated through the cost-benefit analysis model. According to the time series data of total cost and total benefit, the net present value and internal rate of return of the project are calculated, and the cost-benefit analysis results are obtained.

[0110] Collect all necessary input data, including construction cost, operation cost, user distribution and charging pile efficiency; calculate the construction cost of charging piles, including hardware, installation and land costs; estimate the annual operation cost of charging piles, including maintenance, electricity and labor costs; estimate the user time cost based on the user's travel time to the charging pile and the value time cost; predict the potential revenue of charging services based on the utilization rate of charging piles and charging costs; estimate the environmental benefits brought by reduced carbon emissions; add the construction cost, operation cost and user time cost to get the total cost; add the charging service revenue and environmental benefits to get the total benefit; change key parameters such as charging pile utilization, electricity price and other key parameters, and observe their impact on total cost and total benefit; finally output the cost-benefit analysis results, including relevant information such as net present value NPV and internal rate of return IRR.

[0111] By combining the results of cost-benefit analysis with those of accessibility analysis, several charging pile layout schemes are obtained.

[0112] Step 5: Based on several charging pile layout schemes, a corresponding traffic simulation model is established, each charging pile layout scheme is simulated by the traffic simulation model and corresponding simulation results are obtained, and each simulation result is compared to obtain the optimal charging pile layout scheme;

[0113] In this embodiment, based on several charging pile layout schemes, corresponding traffic model parameters and charging pile model parameters are set, and the charging pile layout is dynamically simulated through traffic simulation software and a corresponding traffic simulation model is established;

[0114] Dynamically simulate each charging pile layout plan through the traffic simulation model, collect the corresponding data in each dynamic simulation process, evaluate the performance of each charging pile layout, adjust the location, number or type of charging piles according to the evaluation results, and optimize the charging pile layout plan;

[0115] Import the charging pile layout plan into the traffic simulation model, define the attributes of the charging piles in SUMO, such as charging power, charging type (AC / DC), number of charging piles, set traffic flow and vehicle type, including the type and charging requirements of electric vehicles, define the charging strategy of electric vehicles, charge according to the power level or a fixed time after reaching the charging pile, and set simulation parameters such as simulation duration, time step, etc. Start the SUMO simulation and observe the usage of charging piles and traffic flow in the dynamic simulation. During the simulation, collect key performance indicators, including charging pile utilization, charging waiting time, service level, etc.;

[0116] Conduct sensitivity analysis on the charging pile model parameters in the optimized charging pile layout scheme to observe the impact of different parameter changes on the optimization effect of the charging pile layout scheme; evaluate the effectiveness of the charging pile layout based on the simulation results, and adjust the location, number or configuration of the charging piles according to the situation to optimize the layout; change key parameters such as charging speed, electricity price, user behavior, etc., and observe the impact on the charging pile layout;

[0117] Compare the performance of each optimized charging pile layout solution and select the optimal charging pile layout solution; use SUMO-GU to display the charging pile layout and traffic flow, start the simulation and monitor the usage of the charging piles. Record key data and events during the simulation, analyze the simulation results, and analyze the utilization, service level and user satisfaction of the charging piles. Identify potential bottlenecks in the charging pile layout, and adjust the location, number and configuration of the charging piles based on the analysis results. Rerun the simulation to verify the optimization effect, compare the simulation results of different layout solutions, and evaluate their economy and efficiency. Select the best solution or combine the advantages of multiple solutions;

[0118] Implement the selected charging pile layout plan, establish a monitoring system, collect relevant data of the charging piles and provide feedback, and iteratively optimize the system through the collected charging pile-related data.

[0119] Example 2

[0120] Reference Figure 2 , a system for realizing a method for analyzing the distribution accessibility of charging piles in a city, comprising:

[0121] The data collection and preprocessing module 101 is used to collect relevant data of existing charging piles and historical charging demand data, and perform data preprocessing;

[0122] The data acquisition and preprocessing module 101 includes:

[0123] A data collection submodule is used to obtain road network data, traffic flow data, electric vehicle distribution data, population density data, user historical density data, electric vehicle historical density data and charging pile historical utilization rate data from data sources;

[0124] The data preprocessing submodule is used to clean, integrate and format the collected data to generate charging pile related data sets and historical charging demand data sets.

[0125] A model building module 102, for building a potential location identification model, a demand prediction model and a reachability analysis model based on the preprocessed data;

[0126] The model building module 102 includes:

[0127] Potential location identification submodule: used to identify potential charging pile locations using graph theory and degree centrality evaluation methods;

[0128] The demand forecasting submodule is used to build and optimize the demand forecasting model based on gradient boosting tree and geographically weighted regression to predict the demand for charging piles;

[0129] The accessibility analysis submodule is used to build an accessibility analysis model based on GIS and heuristic site selection algorithm to evaluate the accessibility of charging pile layout.

[0130] The cost-benefit analysis module 103 is used to construct a cost-benefit analysis model and generate multiple charging pile layout plans based on existing data and accessibility analysis results;

[0131] The cost-benefit analysis module 103 includes:

[0132] The cost-benefit analysis framework setting submodule is used to set the basic framework of cost-benefit analysis according to data and requirements;

[0133] The cost-benefit calculation submodule is used to calculate the total cost and total benefit and conduct net present value and internal rate of return analysis;

[0134] The layout scheme generation submodule is used to generate multiple charging pile layout schemes by combining the accessibility analysis results and the cost-benefit analysis results.

[0135] Traffic simulation and optimization module 104, used to establish a traffic simulation model, simulate the traffic conditions under different charging pile layout schemes, and optimize the optimal layout scheme;

[0136] The traffic simulation and optimization module 104 includes:

[0137] Traffic simulation model building submodule, used to build traffic simulation model based on traffic flow theory and simulation software;

[0138] The simulation operation and result analysis submodule is used to simulate the traffic conditions under different charging pile layout schemes and analyze the simulation results;

[0139] The optimization submodule is used to compare the advantages and disadvantages of different layout schemes according to the simulation results and select the optimal charging pile layout scheme.

[0140] The result display and decision support module 105 is used to display the analysis results in a visual form to provide intuitive data support and decision suggestions for decision makers.

[0141] The result display and decision support module 105 includes:

[0142] The result visualization submodule is used to display the charging pile layout plan, accessibility analysis results and cost-benefit analysis results in the form of charts, maps, etc.

[0143] The decision support submodule is used to provide decision makers with specific suggestions for optimizing the layout of charging piles based on the analysis results.

[0144] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for analyzing the distribution accessibility of charging piles in a city, characterized in that: The following steps are involved: Collect relevant data of existing charging piles and historical charging demand data, perform corresponding data preprocessing on the relevant data of existing charging piles and historical charging demand data, and obtain charging pile relevant data sets and historical charging demand data sets respectively; Based on the charging pile related data set and the historical charging demand data set, a potential location identification model and a demand prediction model are constructed respectively, and the charging pile potential location identification results and charging pile demand prediction results are obtained; Construct an accessibility analysis model, combine the potential location identification results of charging piles and the demand prediction results of charging piles, and obtain the accessibility analysis results through the accessibility analysis model; A cost-benefit analysis framework is set up based on the relevant data of existing charging piles and historical charging demand data. A cost-benefit analysis model is constructed based on the cost-benefit analysis framework. Combined with the accessibility analysis results, several charging pile layout plans are obtained. Based on several charging pile layout schemes, a corresponding traffic simulation model is established. Each charging pile layout scheme is simulated through the traffic simulation model and the corresponding simulation results are obtained. Each simulation result is compared to obtain the optimal charging pile layout scheme.

2. The urban charging pile distribution accessibility analysis method according to claim 1, characterized in that: The relevant data of the existing charging piles include road network data, traffic flow data, electric vehicle distribution data and population density data; the historical charging demand data include user historical density data, electric vehicle historical density data and charging pile historical utilization data.

3. The urban charging pile distribution accessibility analysis method according to claim 1, characterized in that: Based on the charging pile related data set, a potential location recognition model is constructed. The specific process of obtaining the potential location recognition result of the charging pile is as follows: Through the charging pile related data set, the starting point, end point, length and speed limit of each road in the road data are obtained; A potential location recognition model is constructed based on graph theory. The starting point and end point of each road are added as nodes to the potential location recognition model, and the length and speed limit of each road are attached as edge attributes. Set the length and speed limit constraints, calculate the edge weights based on the constraints, and update the edge attributes; The importance of each node is evaluated by degree centrality, the average degree centrality of each node is calculated, and nodes with degree centrality greater than the average are selected as potential charging pile locations.

4. The urban charging pile distribution accessibility analysis method according to claim 1, characterized in that: Based on the historical charging demand data set, the demand forecasting model is constructed and the specific process of obtaining the charging pile demand forecasting result is as follows: Extract the basic characteristics and time characteristics of charging demand through the historical charging demand data set, and calculate the corresponding derived characteristics based on the basic characteristics and time characteristics of charging demand; The corresponding demand feature dataset is constructed through the basic features, time features and derived features of charging demand, and the demand feature dataset is divided into training set, validation set and test set according to the time sequence; Based on gradient boosting tree and geographically weighted regression, an initial demand forecasting model is constructed, basic parameters of the initial demand forecasting model are set, the initial demand forecasting model is trained through the training set, the initial demand forecasting model is evaluated through the validation set, model parameters of the initial demand forecasting model are tuned according to the evaluation results and the optimal demand forecasting model is obtained, the optimal demand forecasting model is tested through the test set and the demand forecasting model and the corresponding charging pile demand forecasting results are obtained.

5. The urban charging pile distribution accessibility analysis method according to claim 4, characterized in that: The demand feature dataset includes: Feature set, as input data of demand forecasting model, where each sample contains basic features, time features and derived features of charging demand; The target variable set, which serves as the output data of the demand forecasting model, is a set of target variables, in which each element is associated with a corresponding sample in the feature set.

6. The urban charging pile distribution accessibility analysis method according to claim 1, characterized in that: The specific process of constructing the accessibility analysis model, combining the potential location identification result of the charging pile and the demand prediction result of the charging pile, and obtaining the accessibility analysis result through the accessibility analysis model is as follows: Based on GIS, a reachability analysis model is constructed through Dijkstra algorithm and A* search algorithm; Combining the potential location identification results of charging piles and the demand prediction results of charging piles, the weights of the required number of charging piles, user density and travel time are initialized; Based on the heuristic site selection algorithm, the accessibility analysis results are obtained through the accessibility analysis model.

7. The urban charging pile distribution accessibility analysis method according to claim 1, characterized in that: The specific process of setting up a cost-benefit analysis framework based on the relevant data of existing charging piles and historical charging demand data, constructing a cost-benefit analysis model based on the cost-benefit analysis framework, and obtaining several charging pile layout plans in combination with the accessibility analysis results is as follows: Set up a cost-benefit analysis framework based on the relevant data of existing charging piles and historical charging demand data; Based on the cost-benefit analysis framework, a cost-benefit analysis model is constructed. The total cost and total benefit are calculated through the cost-benefit analysis model. According to the time series data of total cost and total benefit, the net present value and internal rate of return of the project are calculated, and the cost-benefit analysis results are obtained. By combining the results of cost-benefit analysis with those of accessibility analysis, several charging pile layout schemes are obtained.

8. The urban charging pile distribution accessibility analysis method according to claim 1, characterized in that: The specific process of establishing a corresponding traffic simulation model based on several charging pile layout schemes, simulating each charging pile layout scheme through the traffic simulation model and obtaining corresponding simulation results, and comparing each simulation result to obtain the optimal charging pile layout scheme is as follows: Based on several charging pile layout schemes, set the corresponding traffic model parameters and charging pile model parameters, dynamically simulate the charging pile layout through traffic simulation software and establish the corresponding traffic simulation model; Dynamically simulate each charging pile layout plan through the traffic simulation model, collect the corresponding data in each dynamic simulation process, evaluate the performance of each charging pile layout, adjust the location, number or type of charging piles according to the evaluation results, and optimize the charging pile layout plan; Conduct sensitivity analysis on the charging pile model parameters in the optimized charging pile layout scheme to observe the impact of different parameter changes on the optimization effect of the charging pile layout scheme; Compare the performance of each optimized charging pile layout solution and select the best charging pile layout solution; Implement the selected charging pile layout plan, establish a monitoring system, collect relevant data of the charging piles and provide feedback, and iteratively optimize the system through the collected charging pile-related data.

9. A system for implementing the urban charging pile distribution accessibility analysis method according to any one of claims 1 to 8, characterized in that: include: Data collection and preprocessing module, used to collect relevant data of existing charging piles and historical charging demand data, and perform data preprocessing; Model building module, used to build potential location identification model, demand forecasting model and accessibility analysis model based on preprocessed data; Cost-benefit analysis module, which is used to build a cost-benefit analysis model and generate multiple charging pile layout plans based on existing data and accessibility analysis results; Traffic simulation and optimization module, which is used to establish a traffic simulation model, simulate the traffic conditions under different charging pile layout schemes, and optimize the optimal layout scheme; The result display and decision support module is used to display the analysis results in a visual form and provide intuitive data support and decision-making suggestions for decision makers.

10. The urban charging pile distribution accessibility analysis method according to claim 9, characterized in that: The data acquisition and preprocessing module includes: A data collection submodule is used to obtain road network data, traffic flow data, electric vehicle distribution data, population density data, user historical density data, electric vehicle historical density data and charging pile historical utilization rate data from data sources; The data preprocessing submodule is used to clean, integrate and format the collected data to generate charging pile related data sets and historical charging demand data sets; The model building module includes: Potential location identification submodule: used to identify potential charging pile locations using graph theory and degree centrality evaluation methods; The demand forecasting submodule is used to build and optimize the demand forecasting model based on gradient boosting tree and geographically weighted regression to predict the demand for charging piles; The accessibility analysis submodule is used to build an accessibility analysis model based on GIS and heuristic site selection algorithm to evaluate the accessibility of charging pile layout; The cost-benefit analysis module includes: The cost-benefit analysis framework setting submodule is used to set the basic framework of cost-benefit analysis according to data and requirements; The cost-benefit calculation submodule is used to calculate the total cost and total benefit and conduct net present value and internal rate of return analysis; The layout scheme generation submodule is used to generate multiple charging pile layout schemes by combining the accessibility analysis results and the cost-benefit analysis results; The traffic simulation and optimization module includes: Traffic simulation model building submodule, used to build traffic simulation model based on traffic flow theory and simulation software; The simulation operation and result analysis submodule is used to simulate the traffic conditions under different charging pile layout schemes and analyze the simulation results; The optimization submodule is used to compare the advantages and disadvantages of different layout schemes according to the simulation results and select the optimal charging pile layout scheme; The result display and decision support module includes: The result visualization submodule is used to display the charging pile layout plan, accessibility analysis results and cost-benefit analysis results in the form of charts, maps, etc. The decision support submodule is used to provide decision makers with specific suggestions for optimizing the layout of charging piles based on the analysis results.

Images