A charging station site selection optimization method, device, medium and product

Abstract

The present application discloses a charging station site selection optimization method, equipment, medium and product, which relates to the field of charging station site selection. The method includes: constructing a city car charging station site selection model with the goal of minimizing the annual investment cost of the charging station operator, minimizing the user charging cost and maximizing the utilization rate of the charging piles at the charging station; using the information of the proposed charging station as the population particles of the multi-population particle swarm optimization algorithm, using the city car charging station site selection model as the fitness function of the multi-population particle swarm optimization algorithm, iteratively solving the fitness value of each population particle, until the set conditions are met, using the proposed charging station information corresponding to the optimal fitness value as the optimal solution for the charging station site selection. The present application can more accurately and reasonably determine the location of the charging station site with less computing resources, thereby effectively improving the utilization rate of the charging piles during the operation of the charging station.

Description

Technical Field

[0001] The present application relates to the field of charging station site selection, and in particular to a charging station site selection optimization method, equipment, medium and product. Background Art

[0002] In recent years, environmental pollution and energy security issues have become increasingly prominent. As a means of transportation powered by renewable, low-pollution electricity, electric vehicles have become an important choice to replace traditional fuel vehicles. With the gradual increase in the number of electric vehicles, the site selection problem of urban electric vehicle charging stations has become a focus issue that needs to be solved urgently. Reasonable layout of charging stations can effectively reduce the construction cost of charging stations and improve user satisfaction. Therefore, the site selection optimization problem of electric vehicle charging stations is of great significance. There are two main construction ideas of existing charging station site selection models: one is based on deep learning site selection, and the other is based on optimization algorithm to select the site of the existing model.

[0003] For site selection based on deep learning, deep learning methods or new intelligent optimization algorithms are generally used. In site selection based on deep learning, data collection and preprocessing are very critical steps, which directly affect the accuracy of subsequent model training and prediction. In the problem of electric vehicle site selection, data collection involves multiple aspects:

[0004] 1. Geographic location data: This usually includes longitude and latitude coordinates and may also involve geographic divisions at the regional or city level.

[0005] 2. Traffic data: traffic flow, road congestion, location and routes of nearby public transportation stations (such as buses, subways, etc.).

[0006] 3. Demographic and socioeconomic data: population density, income level, consumption habits, vehicle ownership, etc. These data can usually be obtained through censuses, market surveys or government statistical departments.

[0007] 4. Power supply data: grid capacity, power supply stability, electricity prices, etc. These data are crucial for assessing the feasibility of charging station construction.

[0008] 5. Competitor data: The location, quantity, and operation status of nearby charging stations. These data are helpful in analyzing the market competition situation.

[0009] 6. Land and planning data: land prices, land use planning, urban planning, etc. These data have an important impact on assessing the feasibility and cost of site selection.

[0010] Based on the above description, although the site selection method based on deep learning has many advantages, it also has some potential disadvantages, which may affect the performance of the model and the actual application effect. Deep learning models usually have high requirements on the quality and quantity of data. If the collected data is incomplete, inaccurate or biased, the prediction performance of the model will be seriously affected. In some regions or specific situations, relevant data may be difficult to obtain, resulting in limited model training. In addition, site selection problems in different regions may have different characteristics and challenges, and models need to be designed and optimized for specific problems. In addition, when training deep learning models, a large amount of computing resources are usually required, including high-performance central processing units (CPUs), graphics processing units (GPUs), and distributed computing clusters. This may result in longer training time and higher costs. In practical applications, if a large number of candidate locations need to be predicted, higher computing resources may also be required.

[0011] The intelligent optimization algorithm commonly used in site selection for existing models based on optimization algorithms is generally a meta-heuristic algorithm that simulates the social behavior of organisms (including ants, fish, birds, and other organisms in nature). Its main steps include:

[0012] 1. Demand analysis: Determine the main objectives of site selection, such as increasing charging coverage, reducing operating costs, and optimizing user experience.

[0013] 2. Data collection.

[0014] 1) Road network and traffic data: Collect urban road network data and traffic flow data to assess the accessibility and convenience of potential site locations.

[0015] 2) Charging demand and electricity consumption data: Collect charging demand, historical charging data and electricity load data of electric vehicles in the region to predict future charging demand.

[0016] 3) Geographic information data: Collect geographic information data on topography, climate, geology, etc. to evaluate whether the natural conditions of the site are suitable for building a charging station.

[0017] 3. Select appropriate evaluation indicators.

[0018] 4. Build a model and solve it.

[0019] Compared with deep learning methods, the use of metaheuristic algorithms to solve such problems can more quickly find the optimal solution or approximate optimal solution among a large number of potential site selection schemes, thereby improving the efficiency of site selection work. Because it is easy to implement and has simple principles, it is widely used to solve various global problems. However, the method of site selection based on optimization algorithms for existing models also has many disadvantages. For example, the accuracy of the optimization algorithm is highly dependent on the accuracy and completeness of the input data. If the data is biased or missing, it may lead to deviations or unreasonable site selection schemes. In addition, due to the limitations of the algorithm itself or the characteristics of the problem itself, it may fall into a local optimal solution during the solution process and fail to find the global optimal solution. Summary of the invention

[0020] The purpose of this application is to provide a charging station site selection optimization method, equipment, medium and product, which can more accurately and reasonably determine the location of the charging station with smaller computing resources, thereby effectively improving the utilization rate of charging piles during the operation of the charging station.

[0021] To achieve the above objectives, this application provides the following solutions:

[0022] In a first aspect, the present application provides a method for optimizing the site selection of a charging station, comprising:

[0023] A site selection model for urban vehicle charging stations is constructed with the goal of minimizing the annual investment cost for charging station operators, minimizing the charging cost for users, and maximizing the utilization rate of charging piles at charging stations;

[0024] The information of the proposed charging station is used as the population particles of the multi-population particle swarm optimization algorithm, and the urban automobile charging station site selection model is used as the fitness function of the multi-population particle swarm optimization algorithm. The fitness value of each population particle is iteratively solved until the set conditions are met. The proposed charging station information corresponding to the optimal fitness value is used as the optimal solution for the charging station site selection; the proposed charging station information includes the longitude of the proposed charging station, the latitude of the proposed charging station and the number of charging piles of the proposed charging station.

[0025] Optionally, with the goal of minimizing the annual investment cost of charging station operators, minimizing the charging cost of users, and maximizing the utilization rate of charging piles at charging stations, a city car charging station site selection model is constructed, specifically including:

[0026] With the goal of minimizing the annual investment cost of charging station operators, a charging station construction and operation and maintenance cost model for investors is constructed; the constructed charging station construction and operation and maintenance cost model for investors is expressed as:

[0027]

[0028] With the goal of minimizing user charging costs and maximizing the utilization rate of charging piles at charging stations, a user charging cost model is constructed; the constructed user charging cost model is expressed as:

[0029]

[0030] By means of linear weighting, with the goal of minimizing the comprehensive cost of users and charging stations, an investor charging station construction and operation and maintenance cost model is established to complete the construction of the urban automobile charging station site selection model; the established investor charging station construction and operation and maintenance cost model is expressed as:

[0031] F = λ1·F1+λ2·F2;

[0032] In the formula, F1 is the construction and operation and maintenance cost of the charging station by the investor, C1 is the construction cost of the charging station, C2 is the land cost of the charging station, r is the discount rate, n is the service life of the charging station, α is the operation and maintenance cost coefficient, F2 is the user charging cost, I is the number of demand points, and M is the cost of the charging station. i is the charging demand at the demand point, D i,j is the Euclidean distance between demand point i and charging station j, q is the single charging capacity, X i,j is the decision variable, τ is the average waiting time, β is the time cost coefficient of charging users, and Y j is the utilization rate of charging piles in charging station j, λ1 and λ2 are weight coefficients, and F is the comprehensive cost of investors and users.

[0033] Optionally, the process of determining each parameter in the user charging cost model includes:

[0034] The city is divided according to different functional areas to obtain candidate areas for charging stations to be built, and the charging demand M of the demand points in the candidate areas is determined based on the candidate area information. i The candidate area information includes: economic development, regional functional characteristics, number of electric vehicles and population in the area;

[0035] Determine the interval of the number of charging stations to be built in the candidate area and the interval of the number of charging piles in a single charging station according to the urban planning;

[0036] The location of the initial candidate point is determined based on the number interval of charging stations to be built and the number configuration interval of charging piles in a single charging station, and the Delaunay triangulation network is constructed according to the location of the initial candidate point;

[0037] The initial candidate points and the Delaunay triangulation are numbered, and the vertices of each triangle in the Delaunay triangulation are recorded; each vertex is used as a candidate point for the site selection of a charging station; one candidate point corresponds to one charging station;

[0038] Determine and record the circumcenter of each triangle, and determine whether there is an adjacent triangle that shares a common side with the current triangle;

[0039] When there is an adjacent triangle that shares a common edge with the current triangle, connect the circumcenter of the adjacent triangle to the circumcenter of the current triangle and store them in the Voronoi edge list;

[0040] When there is no adjacent triangle sharing a common edge with the current triangle, the edge of the current triangle without the adjacent triangle is used as the side of the Delaunay triangulation, and the perpendicular ray of the side is stored in the Voronoi edge linked list;

[0041] Generating a Voronoi graph according to the Voronoi edge list, and dividing the candidate area according to the Voronoi graph to obtain divided areas;

[0042] The Euclidean distance between the demand point and the candidate point is determined according to the divided area, and this Euclidean distance is used as the Euclidean distance D between the demand point i and the charging station j. i,j ;

[0043] Acquiring charging data, processing the charging data, and taking an average value of charging power of the processed charging data as a single charging power;

[0044] Constructing a scatter plot of the charging date in the processed charging data according to hours, minutes and seconds, fitting the processed charging data into a curve based on the scatter plot, and determining the charging probability distribution of the electric vehicle user in one day according to the fitted curve;

[0045] Based on the charging probability distribution, the Monte Carlo method is used to simulate the charging start time of each charging user within 24 hours, and the average charging time and waiting time of the user are determined according to the single charging power;

[0046] An average waiting time τ is determined based on the queue waiting time.

[0047] Optionally, the information of the proposed charging station is used as the population particles of the multi-population particle swarm optimization algorithm, and the urban automobile charging station site selection model is used as the fitness function of the multi-population particle swarm optimization algorithm. The fitness value of each population particle is iteratively solved until the set conditions are met, and the proposed charging station information corresponding to the optimal fitness value is used as the optimal solution for the charging station site selection, which specifically includes:

[0048] The proposed charging station information is used as the population particles of the multi-population particle swarm optimization algorithm, and the urban automobile charging station site selection model is used as the fitness function of the multi-population particle swarm optimization algorithm to determine the fitness value of each population particle;

[0049] Arrange the population particles in descending order according to the fitness values ​​to obtain a particle sequence;

[0050] The first 50% of the particles in the particle sequence are set as the elite population, the last 50% of the particles in the particle sequence are set as the common population, and the evolution mode of the elite population and the common population is set;

[0051] Iteratively updating the particle speed of the elite population based on the evolutionary method of the elite population, and iteratively updating the particle speed of the ordinary population based on the evolutionary method of the ordinary population;

[0052] The updated particle speed of the elite population and the particle speed of the common population determine the particle position;

[0053] When the particle position stops updating within K consecutive iterations, the position of the particle after jumping out of the local optimal state is obtained based on the current particle position according to the jump-out strategy of reverse learning, and the position of the particle after jumping out of the local optimal state is used as the new particle position;

[0054] Determine whether the number of iterations reaches the set maximum number of iterations;

[0055] When the number of iterations reaches the set maximum number of iterations, the proposed charging station information corresponding to the optimal fitness value is used as the optimal solution for the charging station site selection;

[0056] When the number of iterations does not reach the set maximum number of iterations, return to the step of iteratively updating the particle speed of the elite population based on the evolutionary method of the elite population, and iteratively updating the particle speed of the ordinary population based on the evolutionary method of the ordinary population.

[0057] Optionally, the iterative update formula of the particle velocity of the elite population is:

[0058]

[0059] Where V i',d (t+1) J is the particle velocity of the elite population of the i'th particle in the dth dimension when the iteration number is t+1, V i',d (t) J is the particle speed of the elite population of the i'th particle in the dth dimension when the number of iterations is t, PbestBool(i') is the state value of the optimal fitness value change of the i'th particle, T is the maximum number of iterations set, Pbest i',d is the size of the historical optimal position of the i'th particle in the d dimension, r1 and r2 are random values ​​between [0,1] and [-0.1,0.1], c1 is a fixed constant, w is the inertia weight, is the particle position of the i'th particle in the elite population in the dth dimension when the number of iterations is t.

[0060] Optionally, the iterative update formula of the particle velocity of the common population is:

[0061]

[0062] Where ri is a random integer, Gbest d is the global optimal position, X ri,d is the particle position of the ri-th particle in the ordinary population in the d-th dimension, is the particle position of the i'th particle in the dth dimension of the common population, c1 and c2 are fixed constants, r1 and r2 are random numbers between [0,1], rr is a random integer between [1,i'], rl is a random integer between [1,rr], V i',d (t+1) P is the particle velocity of the i'th particle in the dth dimension when the iteration number is t+1, V i',d (t) P is the particle velocity of the ordinary population of the i'th particle in the dth dimension when the number of iterations is t.

[0063] Optionally, the position of a particle after it jumps out of the local optimum is expressed as:

[0064] x new =r4×Pbest i' +(1-r4)×x ri +r5×(x ri -Pbest i' );

[0065] In the formula, x new Pbest is the position of the particle after it jumps out of the local optimal position. i' is the historical optimal position of particle i', r4 and r5 are both random numbers between [0,1], ri is a random integer between [1,N], x ri is the position of the rith particle, and N is the total number of particles in the population.

[0066] In a second aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the charging station site optimization method provided above.

[0067] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the charging station site optimization method provided above.

[0068] In a fourth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the charging station site optimization method provided above.

[0069] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0070] The present application provides a charging station site selection optimization method, equipment, medium and product. By constructing a city car charging station site selection model with the goal of minimizing the annual investment cost of the charging station operator, minimizing the user charging cost and maximizing the utilization rate of the charging piles at the charging station, it can effectively reduce the investment cost of the charging station investors and improve the charging satisfaction of electric vehicle users. In addition, the constructed city car charging station site selection model comprehensively considers the utilization rate of the charging piles, which can effectively prevent the phenomenon of charging piles being idle for a long time when the charging station is built. By using the city car charging station site selection model as a fitness function and using a multi-swarm particle swarm optimization algorithm to determine the optimal solution for the charging station site selection, it can effectively suppress the problems of premature maturity and easy to fall into local optimality in the iterative process of the particle swarm optimization algorithm, effectively improve the search range and search accuracy of the population, and more accurately and reasonably determine the location of the charging station with smaller computing resources, thereby effectively improving the utilization rate of the charging piles during the operation of the charging station. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0072] Figure 1 This is an application environment diagram of a charging station site selection optimization method in an embodiment of the present application;

[0073] Figure 2 A schematic diagram of a flow chart of a charging station site selection optimization method provided in one embodiment of the present application;

[0074] Figure 3 A flowchart of implementing a multi-swarm particle swarm optimization algorithm provided in an embodiment of the present application;

[0075] Figure 4 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0076] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0077] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0078] The charging station site selection optimization method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. The terminal 102 can send the information of the proposed charging site to the server 104. After the server 104 receives the information of the proposed charging site, for the information of the proposed charging site, the server 104 takes the annual investment cost of the charging station operator, the user charging cost and the charging station charging pile utilization rate as the goal, and constructs a city car charging station site selection model. The proposed charging site information is used as the population particle of the multi-swarm particle swarm optimization algorithm, and the urban car charging station site selection model is used as the fitness function of the multi-swarm particle swarm optimization algorithm. The fitness value of each population particle is iteratively solved until the set condition is met, and the proposed charging site information corresponding to the optimal fitness value is used as the optimal solution for the charging station site selection. The server 104 can feedback the obtained optimal solution for the charging station site selection to the terminal 102. In addition, in some embodiments, the charging station site selection optimization method can also be implemented independently by the server 104 or the terminal 102. For example, the terminal 102 can directly determine the optimal solution for the charging station site selection based on the information of the proposed charging station to complete the charging station site selection optimization. The server 104 can also obtain the information of the proposed charging station from the data storage system and determine the optimal solution for the charging station site selection.

[0079] The terminal 102 may be, but is not limited to, various desktop computers, laptop computers, smart phones, tablet computers, IoT devices, and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart vehicle-mounted devices, etc. The portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers, or may be a cloud server.

[0080] In an exemplary embodiment, Figure 2 As shown, a charging station location optimization method is provided. The method is executed by a computer device, and can be executed by a computer device such as a terminal or a server alone, or can be executed by a terminal and a server together. In the embodiment of the present application, the method is applied to Figure 1 The server 104 in the example is used as an example to illustrate, including the following steps 200 to 201. Among them:

[0081] Step 200: Construct a city vehicle charging station site selection model with the goal of minimizing the annual investment cost of the charging station operator, minimizing the user charging cost, and maximizing the utilization rate of charging piles at the charging station.

[0082] Step 201: The proposed charging station information is used as the population particles of the multi-population particle swarm optimization algorithm, and the urban automobile charging station site selection model is used as the fitness function of the multi-population particle swarm optimization algorithm. The fitness value of each population particle is iteratively solved until the set conditions are met, and the proposed charging station information corresponding to the optimal fitness value is used as the optimal solution for the charging station site selection. The proposed charging station information includes the longitude of the proposed charging station, the latitude of the proposed charging station, and the number of charging piles of the proposed charging station.

[0083] By implementing the above steps 200 to 201, the present application establishes a more complete target constraint model (i.e., the urban automobile charging station site selection model), and uses the improved multi-population particle swarm optimization algorithm to solve the model, making the solution of the objective function more scientific and reasonable. Compared with the traditional electric vehicle charging station site selection model, the present application proposes new constraints for the idle charging piles in the current charging stations, which can effectively improve the utilization rate of charging piles in the operation of the charging station, and significantly reduce the operating costs of enterprises and improve the return on investment of charging station operators in the future trend of electric vehicles replacing fuel-consuming vehicles.

[0084] In another exemplary embodiment of the present application, in order to effectively reduce the investment cost of charging station investors and improve the charging satisfaction of electric vehicle users, and to effectively prevent the phenomenon of charging piles being idle for a long time when building charging stations, the above step 200 is replaced by the following steps 300 to 302:

[0085] Step 300: With the goal of minimizing the annual investment cost of the charging station operator, construct an investor charging station construction and operation and maintenance cost model. The annual operator investment cost includes the charging station construction cost and the operation and maintenance cost. The constructed investor charging station construction and operation and maintenance cost model is expressed as:

[0086]

[0087] Step 301: With the goal of minimizing the user charging cost and maximizing the utilization rate of charging piles at charging stations, a user charging cost model is constructed. The constructed user charging cost model is expressed as:

[0088]

[0089] Step 302: By means of linear weighting, with the goal of minimizing the comprehensive cost of users and charging stations, an investor charging station construction and operation and maintenance cost model is established to complete the construction of the urban automobile charging station site selection model. The established investor charging station construction and operation and maintenance cost model is expressed as:

[0090] F=λ1·F1+λ2·F2 (3)

[0091] In the formula, F1 is the construction and operation and maintenance cost of the charging station by the investor, C1 is the construction cost of the charging station, C2 is the land cost of the charging station, r is the discount rate, n is the service life of the charging station, α is the operation and maintenance cost coefficient, F2 is the user charging cost, I is the number of demand points, and M is the cost of the charging station. i is the charging demand at the demand point, D i,j is the Euclidean distance between demand point i and charging station j, q is the single charging capacity, X i,j is the decision variable, τ is the average waiting time, β is the time cost coefficient of the charging user, Y is the utilization rate of the charging piles in the charging station j, λ1 and λ2 are weight coefficients, and F is the comprehensive cost of investors and users.

[0092] In another exemplary embodiment of the present application, in order to effectively suppress the problems of premature maturity and easy falling into local optimality in the iteration process of the particle swarm optimization algorithm, effectively improve the search range and search accuracy of the population, and more accurately and reasonably determine the location of the charging station with smaller computing resources, thereby effectively improving the utilization rate of charging piles during the operation of the charging station, the particle swarm optimization algorithm is improved to obtain a multi-swarm particle optimization algorithm. Based on this, the above step 201 is replaced by the following steps 400 to 408:

[0093] Step 400: Using the information of the proposed charging stations as the population particles of the multi-population particle swarm optimization algorithm, using the urban automobile charging station site selection model as the fitness function of the multi-population particle swarm optimization algorithm, and determining the fitness value of each population particle.

[0094] In actual application, the multi-swarm particle swarm optimization algorithm can be initialized first. For example, the total number of particles in all populations is set to N = 40, the maximum number of iterations of the population is set to T = 1000, and p proposed sites are randomly selected (p is the number of proposed sites). The longitude, latitude, and number of charging piles corresponding to the proposed sites are used as the input of the multi-swarm particle swarm optimization algorithm. Each particle in the population represents a candidate solution for the location of a charging station. For example, the information of the proposed charging station (i.e., the candidate solution) is recorded as {Y1, Y2, ..., Y p}, where Y1, Y2, Y p They represent the information of the planned charging stations with corresponding numbers, including the longitude, latitude and number of charging piles of the station.

[0095] Step 401: Arrange the population particles in descending order according to the fitness values ​​to obtain a particle sequence.

[0096] Step 402: Divide the particle swarm into two populations, for example, set the first 50% of the particles in the particle sequence as the elite population, set the last 50% of the particles in the particle sequence as the common population, and set the evolutionary methods of the elite population and the common population.

[0097] Step 403: Iteratively update the particle velocity of the elite population based on the evolutionary method of the elite population, and iteratively update the particle velocity of the ordinary population based on the evolutionary method of the ordinary population.

[0098] In the actual application process, different evolutionary methods are designed according to the corresponding functions of the two populations. Among them, the elites in the population (i.e., the elite population) should take the responsibility of exploring the search area as widely as possible, rather than gathering in a small area. Considering that the best position Pbest of the elite particles has good characteristics, they may be close to the local optimum or even the global optimum. Therefore, the elite particles should weaken the traction of the global extreme value Gbest, become more confident, and increase more self-cognition. Based on this, the iterative update formula of the particle speed of the elite population is:

[0099]

[0100] Where V i',d (t+1) J is the particle velocity of the elite population of the i'th particle in the dth dimension when the iteration number is t+1, V i',d (t) J is the particle velocity of the elite population of the ith particle in the dth dimension when the number of iterations is t. PbestBool(i') is the state value of the optimal fitness value change of the ith particle, updated to 1, otherwise 0. T is the maximum number of iterations set, Pbest i',d is the size of the historical optimal position of the i'th particle in the d dimension, r1 and r2 are random values ​​between [0,1] and [-0.1,0.1], c1 is a fixed constant, w is the inertia weight, is the particle position of the i'th particle in the elite population in the dth dimension when the number of iterations is t.

[0101] It is unrealistic to use particles in the ordinary population to explore the solution space because their solution quality is not as good as that of particles in the elite population. The main responsibility of particles in the ordinary population is to assist particles in the elite population in solving problems and improve the diversity of the population. This application proposes a random mean learning strategy. Based on this, the iterative update formula of the particle velocity of the ordinary population is:

[0102]

[0103] Where ri is a random integer, Gbest d is the global optimal position, X ri，d is the particle position of the ri-th particle in the ordinary population in the d-th dimension, is the particle position of the i′th particle in the ordinary population in the dth dimension, c1 and c2 are fixed constants, r1 and r2 are random numbers between [0,1], rr is a random integer between [1,i′], rl is a random integer between [1,rr], V i′，d (t+1) P is the particle velocity of the i'th particle in the dth dimension when the iteration number is t+1, V i′，d (t) P is the particle velocity of the ordinary population of the i'th particle in the dth dimension when the number of iterations is t.

[0104] Step 404: The particle position is determined by the updated particle speed of the elite population and the particle speed of the common population. The particle position is updated according to the updated particle speed. The specific formula is as follows:

[0105] X i' (t+1)=V i' (t+1)+X i' (t) (6)

[0106] In the formula, X i′ (t+1) is the position of the i'th particle at the t+1th iteration, V i (t+1) is the velocity of the i'th particle at the t+1th iteration, X i′ (t) is the position of the i'th particle at the tth iteration.

[0107] Step 405: When the particle position stops updating within K consecutive iterations, the position of the particle after jumping out of the local optimum is obtained based on the current particle position according to the jump-out strategy of reverse learning, and the position of the particle after jumping out of the local optimum is used as the new particle position.

[0108] The purpose of step 405 is to prevent particles from falling into the local optimum under certain circumstances, causing the particles to stop evolving within multiple generations. At this time, in order to make the particles jump out of the local optimum, when the particles stop updating within K (for example, K=10) generations, they jump out of the current position according to the jump-out strategy of reverse learning. The position of the particle after jumping out of the local optimum is expressed as:

[0109] x new =r4×Pbest i' +(1-r4)×x ri +r5×(x ri-Pbest i' ) (7)

[0110] In the formula, x new Pbest is the position of the particle after it jumps out of the local optimal position. i′ is the historical optimal position of particle i', r4 and r5 are both random numbers between [0,1], ri is a random integer between [1,N], x ri is the position of the rith particle, and N is the total number of particles in the population.

[0111] Step 406: Determine whether the number of iterations reaches the set maximum number of iterations.

[0112] Step 407: When the number of iterations reaches the set maximum number of iterations, the proposed charging station information corresponding to the optimal fitness value is used as the optimal solution for charging station site selection.

[0113] Step 408: When the number of iterations does not reach the set maximum number of iterations, return to step 403.

[0114] Based on the above description, the actual solution process of the multi-swarm particle swarm optimization algorithm can be found in Figure 3 The process shown will not be repeated again.

[0115] In another exemplary embodiment of the present application, during the actual application of the urban vehicle charging station site selection model, it is necessary to first determine the specific values ​​of its parameters, for example:

[0116] Step 500: Determine charging demand points and demand quantities.

[0117] The city is divided according to different functional areas, and the geometric center of the divided area is used as the charging demand point, and the latitude and longitude of the demand point is recorded (the characteristics of the demand point are large flow of people and high probability of user charging demand). The charging demand M of the demand point in the selected area is determined by considering the economic development of the region, regional functional characteristics, electric vehicle ownership, and population in the region. i .

[0118] Step 501: Determine the number interval of charging stations to be built in the candidate area and the number interval of charging piles in a single charging station according to the urban planning.

[0119] Step 502: Initialize the charging station site selection model, construct the Delaunay triangulation according to the candidate point positions, number the candidate points and the formed triangles, and record which three candidate points each triangle is composed of. Then calculate and record the circumcenter of each triangle, and finally traverse the triangle list to find the adjacent triangles A, B, and C that share three sides with the current triangle. If it can be found, the circumcenter of the found triangle is connected to the circumcenter of the current triangle and stored in the Voronoi edge list. If it cannot be found, it means that there is no triangle adjacent to the edge, that is, the edge is the outermost edge of the Delaunay triangulation, and the perpendicular ray of the edge is calculated and stored in the Voronoi edge list. After the traversal is completed, the Voronoi diagram is drawn according to the Voronoi edge list, and the urban area where the charging station is to be built is divided according to the Voronoi diagram to obtain the divided area.

[0120] Step 503: Determine the Euclidean distance between the demand point and the candidate point according to the divided area, and use this Euclidean distance as the Euclidean distance between the demand point i and the charging station j, that is, the distance D from the demand point to the corresponding charging station in formula (2) i,j to update.

[0121] Step 504: acquiring charging data, processing the charging data, and taking the average value of the charging power of the processed charging data as the single charging power.

[0122] For example, 102,000 charging data are collected, and the data content includes: 1) Charging date. 2) Charging time length. 3) Longitude of the charging pile. 4) Latitude of the charging pile. 5) The amount of charge for this charging. In order to make the collected data more representative, the collected data needs to be preprocessed. Since the charging time length and charging amount can reflect whether there is an abnormality in this charging, in order to eliminate the abnormal data in the collected data, the charging amount in the data is selected as the evaluation index, and the maximum value, minimum value, median, upper quartile and lower quartile of the charging amount are determined. According to the above five indicators, the corresponding box plot is drawn to analyze the characteristics of the data distribution. According to the box plot, the values ​​of charging power less than QL-1.5IQR are eliminated, and the average value of the charging power of the remaining data is taken as the single charging power q in F2.

[0123] Step 505: construct a scatter plot of the charging date in the processed charging data according to hours, minutes and seconds, and fit the processed charging data into a curve based on the scatter plot, and determine the charging probability distribution of the electric vehicle user in one day according to the fitted curve.

[0124] Step 506: Based on the charging probability distribution, the Monte Carlo method is used to simulate the charging start time (i.e., the charging user's arrival time) of each charging user within 24 hours. If there is an idle charging pile, charging starts immediately, and the charging waiting time is recorded as 0. If there is no idle charging pile at present, waiting is carried out, and the average charging time of the user is determined based on the single charging power, and the queue waiting time is calculated according to the first-come-first-charge rule.

[0125] Step 507: Determine the average waiting time τ based on the queue waiting time. In actual application, the simulation is repeated multiple times to obtain a large number of samples, and the charging waiting time of all users in the simulation process is statistically calculated. Finally, the average of the statistical data is calculated to obtain the average waiting time τ in formula (2).

[0126] In another exemplary embodiment of the present application, taking the use of Matlab 2021a programming software and the visualization tool plot to achieve charging station site optimization as an example, the implementation process of the charging station site optimization method provided by the present application is described in detail. Specifically:

[0127] Step 1: Process the collected data based on Matlab 2021a programming software and visualization tool plot, and design, calculate and simulate according to the established model and multi-swarm particle swarm optimization algorithm.

[0128] Step 2: Collect charging data from previous charging stations, which records the charging station information and the charging information of electric vehicle charging users in detail. The required information includes: the amount of charging, the length of time spent on charging, the start time of charging, and the longitude and latitude of the charging equipment used.

[0129] Step 3: To ensure the authenticity and rationality of the data, draw a box plot based on the charged capacity, eliminate the values ​​with a charged capacity less than QL-1.5IQR based on the box plot, and calculate the average charged capacity of the remaining data.

[0130] Step 4: Use the processed data to construct a scatter plot of the charging dates in the data according to hours, minutes and seconds, and fit it into a curve, and determine the charging probability distribution of electric vehicle users in a day based on the fitted curve.

[0131] Step 5: Determine the Euclidean distance between the demand point and the corresponding charging station candidate point based on the above-mentioned divided areas.

[0132] Step 6: Initialize the parameters of the city location model, including the single charge power q, investment period n, annual discount rate α, and daily charging behavior probability function of electric vehicle users in the city location model. Initialize the parameters of the multi-population particle swarm optimization algorithm. The parameters of the multi-population particle swarm optimization algorithm include the number of particle populations and the maximum number of iterations.

[0133] Step 7: Use the constructed urban charging station site selection model as the fitness function, calculate the fitness value of each individual in the population, and sort the particles according to the fitness value. Set the top 50% of the particles in fitness as the elite population, and the bottom 50% as the ordinary population, and determine the behavior pattern of the particles according to different populations.

[0134] Step 8. Since the ordinary population ranks low in fitness, it is responsible for expanding the search range of the population and improving the quality of the particles in the population as auxiliary evolution particles. When the random number ri is not equal to the current ranking of the particle, the particle randomly selects the random mean strategy for learning. Otherwise, the particle directly selects the best particle in the population for learning.

[0135] Step 9: The elite population is mainly responsible for improving the convergence accuracy and expanding the search range as much as possible. Therefore, two behavior modes are designed for the elite population. When the individual optimal position of the previous generation of particles changes, it means that the current area can be further explored, that is, the particle selects the random oscillation strategy to update the particle speed: Otherwise, the self-learning strategy of learning from the historical optimal position of the particle is selected to update the particle velocity: V i′，d (t+1)=w·V i′，d (t)+c1·r1·(Pbest i′，d -X i′，d ).

[0136] Step 10: Determine whether all particles in the population are trapped in the local optimum. When the historical fitness value of a particle is not updated within K = 10 generations, it is considered that the particle is trapped in the local optimum, and the jump-out strategy is used to update the particle. The update formula is shown in formula (6).

[0137] Step 11: Determine whether the current number of iterations is less than the maximum number of iterations. If the current number of iterations is greater than or equal to the maximum number of iterations, exit the loop and output the optimal particle position and its fitness value. Otherwise, return to step 7 and continue to execute until the algorithm end condition is met.

[0138] Based on the above description, compared with the prior art, the technical solution provided by this application has the following advantages:

[0139] (1) This application constructs a city charging station site selection model based on minimizing the cost of charging investors and charging users, and comprehensively considering factors such as the utilization rate of charging piles, which can effectively reduce the investment cost of charging station investors and improve the charging satisfaction of electric vehicle users. In addition, the constructed model comprehensively considers the utilization rate of charging piles, which can effectively prevent the phenomenon of charging piles being idle for a long time when building charging stations.

[0140] (2) This application proposes the idea of ​​dividing the multi-population particle swarm optimization algorithm into different classes, and proposes a random mean learning strategy, which can effectively suppress the problems of premature maturity and easy falling into local optimality in the iterative process of the particle swarm optimization algorithm, and can effectively improve the search range and search accuracy of the population.

[0141] (3) This application uses actual charging station charging data as raw data and performs preprocessing to make the model construction more scientific and reasonable.

[0142] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store charging station site optimization data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a charging station site optimization method is implemented.

[0143] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0144] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0145] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0146] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0147] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0148] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0149] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0150] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A charging station site selection optimization method, characterized in that: include: A site selection model for urban vehicle charging stations is constructed with the goal of minimizing the annual investment cost for charging station operators, minimizing the charging cost for users, and maximizing the utilization rate of charging piles at charging stations; The proposed charging station information is used as the population particles of the multi-population particle swarm optimization algorithm, and the urban automobile charging station site selection model is used as the fitness function of the multi-population particle swarm optimization algorithm to determine the fitness value of each population particle; Arrange the population particles in descending order according to the fitness value to obtain the particle sequence; Set the first 50% of particles in the particle sequence as the elite population, set the last 50% of particles in the particle sequence as the common population, and set the evolution mode of the elite population and the common population; Iteratively update the particle speed of the elite population based on the evolutionary method of the elite population, and iteratively update the particle speed of the ordinary population based on the evolutionary method of the ordinary population; Determine the particle position based on the updated particle velocity of the elite population and the particle velocity of the common population; When the particle position stops updating within K consecutive iterations, the position of the particle after jumping out of the local optimal state is obtained based on the current particle position according to the jump-out strategy of reverse learning, and the position of the particle after jumping out of the local optimal state is used as the new particle position; Determine whether the number of iterations reaches the set maximum number of iterations; When the number of iterations reaches the set maximum number of iterations, the proposed charging station information corresponding to the optimal fitness value is used as the optimal solution for the charging station site selection; When the number of iterations does not reach the set maximum number of iterations, the steps of iteratively updating the particle speed of the elite population based on the evolutionary method of the elite population and iteratively updating the particle speed of the ordinary population based on the evolutionary method of the ordinary population are returned; the information of the proposed charging station includes the longitude of the proposed charging station, the latitude of the proposed charging station and the number of charging piles of the proposed charging station; The iterative update formula of the particle velocity of the elite population is: In the formula, is the particle velocity of the elite population of the i'th particle in the dth dimension when the iteration number is t+1, is the particle speed of the elite population of the i'th particle in the dth dimension when the number of iterations is t, PbestBool(i') is the state value of the optimal fitness value change of the i'th particle, T is the maximum number of iterations set, Pbest i',d is the size of the historical optimal position of the i'th particle in the d dimension, r1 and r2 are random values ​​between [0,1] and [-0.1,0.1], c1 is a fixed constant, w is the inertia weight, is the particle position of the i'th particle in the elite population in the dth dimension when the number of iterations is t; The iterative update formula of the particle velocity of the common population is: Where ri is a random integer between [1, N], Gbest d is the global optimal position, X ri,d is the particle position of the ri-th particle in the ordinary population in the d-th dimension, is the particle position of the i'th particle in the dth dimension of the common population, c1 and c2 are fixed constants, r3 and r6 are random numbers between [0,1], rr is a random integer between [1,i'], rl is a random integer between [1,rr], V i',d (t+1) P is the particle velocity of the i'th particle in the dth dimension when the iteration number is t+1, V i',d (t) P is the particle velocity of the i'th particle in the ordinary population in the dth dimension when the number of iterations is t, and N is the total number of particles in the population.

2. The method for optimizing the location of a charging station according to claim 1, characterized in that: With the goal of minimizing the annual investment cost of charging station operators, minimizing the charging cost of users, and maximizing the utilization rate of charging piles at charging stations, a site selection model for urban automobile charging stations is constructed, which specifically includes: With the goal of minimizing the annual investment cost of charging station operators, a charging station construction and operation and maintenance cost model for investors is constructed; the constructed charging station construction and operation and maintenance cost model for investors is expressed as: With the goal of minimizing user charging costs and maximizing the utilization rate of charging piles at charging stations, a user charging cost model is constructed; the constructed user charging cost model is expressed as: By means of linear weighting, with the goal of minimizing the comprehensive cost of users and charging stations, an investor charging station construction and operation and maintenance cost model is established to complete the construction of the urban automobile charging station site selection model; the established investor and user comprehensive cost model is expressed as: F = λ1·F1+λ2·F2; In the formula, F1 is the construction and operation and maintenance cost model of the investor's charging station, C1 is the construction cost of the charging station, C2 is the land cost of the charging station, r is the discount rate, n is the service life of the charging station, α is the operation and maintenance cost coefficient, F2 is the user charging cost model, I is the number of demand points, M i is the charging demand at the demand point, D i,j' is the Euclidean distance between demand point i and charging station j', q is the single charging capacity, X i,j' is the decision variable, τ is the average waiting time, β is the time cost coefficient of charging users, and Y j' is the utilization rate of charging piles in charging station j', λ1 and λ2 are weight coefficients, and F is the comprehensive cost model for investors and users.

3. The method for optimizing the location of a charging station according to claim 2, characterized in that: The process of determining each parameter in the user charging cost model includes: The city is divided according to different functional areas to obtain candidate areas for charging stations to be built, and the charging demand M of the demand points in the candidate areas is determined based on the candidate area information. i The candidate area information includes: economic development, regional functional characteristics, number of electric vehicles and population in the area; Determine the interval of the number of charging stations to be built in the candidate area and the interval of the number of charging piles in a single charging station according to the urban planning; The location of the initial candidate point is determined based on the number interval of charging stations to be built and the number configuration interval of charging piles in a single charging station, and the Delaunay triangulation network is constructed according to the location of the initial candidate point; The initial candidate points and the Delaunay triangulation are numbered, and the vertices of each triangle in the Delaunay triangulation are recorded; each vertex is used as a candidate point for the site selection of a charging station; one candidate point corresponds to one charging station; Determine and record the circumcenter of each triangle, and determine whether there is an adjacent triangle that shares a common side with the current triangle; When there is an adjacent triangle that shares a common edge with the current triangle, connect the circumcenter of the adjacent triangle to the circumcenter of the current triangle and store them in the Voronoi edge list; When there is no adjacent triangle sharing a common edge with the current triangle, the edge of the current triangle without the adjacent triangle is used as the side of the Delaunay triangulation, and the perpendicular ray of the side is stored in the Voronoi edge linked list; Generating a Voronoi graph according to the Voronoi edge list, and dividing the candidate area according to the Voronoi graph to obtain divided areas; The Euclidean distance between the demand point and the candidate point is determined according to the divided area, and this Euclidean distance is used as the Euclidean distance D between the demand point i and the charging station j' i,j' ; Acquiring charging data, processing the charging data, and taking an average value of charging power of the processed charging data as a single charging power; Constructing a scatter plot of the charging date in the processed charging data according to hours, minutes and seconds, fitting the processed charging data into a curve based on the scatter plot, and determining the charging probability distribution of the electric vehicle user in one day according to the fitted curve; Based on the charging probability distribution, the Monte Carlo method is used to simulate the charging start time of each charging user within 24 hours, and the average charging time and waiting time of the user are determined according to the single charging power; An average waiting time τ is determined based on the queue waiting time.

4. The method for optimizing the location of a charging station according to claim 1, characterized in that: The position of the particle after jumping out of the local optimum is expressed as: x new =r4×Pbest i' +(1-r4)×x ri +r5×(x ri -Pbest i' ); In the formula, x new Pbest is the position of the particle after it jumps out of the local optimal position. i' is the historical optimal position of particle i', r4 and r5 are both random numbers between [0,1], x ri is the position of the rith particle.

5. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the charging station location optimization method according to any one of claims 1 to 4.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for optimizing the location of a charging station according to any one of claims 1 to 4 is implemented.

7. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for optimizing the location of a charging station according to any one of claims 1 to 4 is implemented.

Images