Electric vehicle charging station planning method based on load and traffic flow distribution

Abstract

The invention relates to an electric vehicle charging station planning method based on load and traffic flow distribution, and the method considers the superposition influence of topographic data, historical load data and historical traffic flow data in planning, and carries out the prediction of the fixed load distribution of a charging station and the charging load of an electric vehicle through the historical load data. In combination with the analysis of a prediction result and topographic data, a power distribution network grid map is optimized by adopting a power distribution optimization model, and then grid region load division is performed, so that the influence of the topographic data and historical load data on the grid planning layout of the charging station can be described more accurately; meanwhile, a grid area traffic heat distribution condition is analyzed by combining a power distribution network grid optimization map and historical traffic flow data, and finally, the superposition influence of grid area load division and grid area traffic heat division results is comprehensively considered to determine a capacity planning scheme of the electric vehicle charging pile. Therefore, the accuracy and rationality of electric vehicle charging station planning can be better ensured.

Description

Technical Field

[0001] The present invention relates to the fields of power system analysis technology and power distribution network technology, and in particular to a method for planning electric vehicle charging stations based on load and vehicle flow distribution. Background Art

[0002] Charging facility planning refers to the systematic arrangement and design process of providing charging services for electric vehicles in a specific area. This process involves considerations such as determining the location of charging stations, the number of charging piles, and the construction of related infrastructure. The goal of charging facility planning is usually to meet the charging needs of electric vehicles, increase the coverage of charging facilities, and promote the development of electric transportation.

[0003] Charging station planning is a systematic process that requires comprehensive consideration of multiple factors such as demand, location, power supply, user experience, etc. to ensure that charging facilities can meet user needs, achieve sustainable development, and comply with local environment and regulations. The purpose of charging station facility planning is mainly to ensure that electric vehicles can obtain effective, convenient and sustainable charging services, and to establish a reliable, efficient and sustainable charging infrastructure to promote the development of electric transportation and meet the growing market demand for electric vehicles. At the same time, the planning of charging station facilities is highly uncertain, such as the uncertainty of user-side load demand, the uncertainty of power supply, and the uncertainty of the location of charging stations, which further increases the difficulty of solving the problem of charging station facility planning.

[0004] The technology of charging station facility planning is constantly developing, involving multiple fields, usually considering geographic information factors, historical data and forecast data, power system analysis, etc. However, the degree of coupling between the transportation network and the power transmission network is not high enough, which will lead to problems such as long planning and construction cycle of charging station facilities and insufficient rationality of charging station facility planning. Therefore, how to design a charging pile location and capacity planning scheme for urban electric vehicle charging stations that considers multiple factors to improve the efficiency and rationality of its implementation planning, and solve its planning difficulties, long planning cycle, and insufficient planning rationality. It has become a technical direction that needs to be studied urgently. Summary of the invention

[0005] In view of the deficiencies in the above-mentioned prior art, the present invention provides an electric vehicle charging station planning method based on load and traffic flow distribution, so as to improve the efficiency and rationality of electric vehicle charging station planning, thereby solving the problems in the prior art of difficult planning of electric vehicle charging stations in urban areas with complex terrain, long planning cycle and insufficient planning rationality.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0007] A method for planning electric vehicle charging stations based on load and traffic flow distribution includes the following steps:

[0008] S1. Obtain terrain data, historical load data and historical vehicle flow data of the area to be planned; the terrain data includes slope information and a plane-based distribution network grid map; the historical load data includes fixed load demand data and electric vehicle charging load demand data;

[0009] S2. Predicting the fixed load and the electric vehicle charging load according to the historical load data to obtain a fixed load distribution prediction result and an electric vehicle charging load prediction result;

[0010] S3, according to the slope information, the distribution network grid map, and the obtained fixed load distribution prediction results and electric vehicle charging load prediction results, a preset distribution optimization model is used to optimize the distribution network grid optimization map with total cost minimization as the objective function;

[0011] S4. According to the fixed load distribution prediction result and the electric vehicle charging load prediction result, the grid area in the distribution network grid optimization map is divided into load levels to obtain a grid area load division result;

[0012] S5. Based on the historical vehicle flow data, the grid areas in the distribution network grid optimization map are divided into traffic thermal levels to obtain grid area traffic thermal division results;

[0013] S6. Obtaining a capacity planning scheme for electric vehicle charging piles in different grid areas in the area to be planned based on the grid area load division result and the grid area traffic thermal division result;

[0014] S7. According to the obtained electric vehicle charging pile capacity planning scheme, the capacity of the electric vehicle charging piles distributed in different grid areas in the planning area is planned.

[0015] As a preferred solution, step S2 is specifically as follows:

[0016] S201, performing internal model function decomposition on the fixed load demand data and the electric vehicle charging load demand data in the historical load data, respectively, to obtain fixed load internal model function components and electric vehicle charging load internal model function components;

[0017] S202, using the ensemble empirical mode decomposition method to decompose each fixed load internal mode function component and the electric vehicle charging load internal mode function component, respectively, to obtain a decomposed fixed load time series and an electric vehicle charging load time series;

[0018] S203, according to the decomposed fixed load time series and electric vehicle charging load time series, determine the information entropy IE of each fixed load internal model function component and electric vehicle charging load internal model function component, and reconstruct the fixed load extended internal model function component and electric vehicle charging load extended internal model function component with similar information entropy IE; the similar information entropy IE means that the difference between the information entropy IE of the two is within a preset difference range;

[0019] S204, using the information entropy IE of the fixed load internal model function component and the fixed load extended internal model function component as the input of the preset LSTM model to predict, and obtain the first fixed load distribution prediction result; using the information entropy IE of the electric vehicle charging load internal model function component and the electric vehicle charging load extended internal model function component as the input of the LSTM model to predict, and obtain the first electric vehicle charging load prediction result;

[0020] S205, using the information entropy IE of the fixed load internal model function component and the fixed load extended internal model function component as the input of the preset GMDH model to predict, and obtain a second fixed load distribution prediction result; using the information entropy IE of the electric vehicle charging load internal model function component and the electric vehicle charging load extended internal model function component as the input of the GMDH model to predict, and obtain a second electric vehicle charging load prediction result;

[0021] S206. Perform a weighted summation on the first fixed load distribution prediction result and the second fixed load distribution prediction result to obtain a final fixed load distribution prediction result; perform a weighted summation on the first electric vehicle charging load prediction result and the second electric vehicle charging load prediction result to obtain a final electric vehicle charging load prediction result.

[0022] As a preferred solution, step S206 is specifically as follows:

[0023] First, the first weight value of the weighted sum of the first fixed load distribution prediction result and the second fixed load distribution prediction result, and the second weight value of the weighted sum of the first electric vehicle charging load prediction result and the second electric vehicle charging load prediction result are optimized using the bitter fish optimization algorithm, and then the optimized first weight value and second weight value are used to weighted sum and calculate the fixed load distribution prediction result and the electric vehicle charging load prediction result.

[0024] As a preferred solution, in step S206, for any one of the first weight value and the second weight value, the specific method of using the bitter fish optimization algorithm for optimization is:

[0025] S2061, randomly generate n initial weight values ​​as bitter fish individuals of the bitter fish population, i = 1, 2, ..., n, and obtain the initialized bitter fish population Pop, that is:

[0026] Pop=[F i |i=1,2,…,n],F i =[f i,k |k=1,2,…,m];

[0027] Among them, F i ∈Pop is any i-th bitter fish individual in the bitter fish population, and the individuals in the bitter fish population are the weight values ​​to be optimized; f i,k is any i-th bitter fish individual F in the bitter fish population i The value in the kth dimension; n is the number of bitter fish individuals contained in the bitter fish population in the bitter fish optimization algorithm, and m is the dimension of each bitter fish individual in the bitter fish population;

[0028] Then, the number of iterations is initialized to t=1, and the process goes to step S2062;

[0029] S2062. Randomly divide half of the bitter fish individuals in the bitter fish population into positive individuals and mark them as F. i (+), and the other half of the bitter fish individuals are negative individuals and are marked as F j (-);

[0030] S2063, for each positive individual F i (+), iterative update is performed in the following way:

[0031]

[0032] In the formula, and They are positive individuals F i (+) Positions in the tth and t+1th iterations; is a randomly selected oyster position; is the optimal solution of the bittern in the tth iteration, that is, the bittern closest to the oyster in the tth iteration; δ and r are random numbers between 0 and 1; J(t) is the number of steps or speed that the bittern moves to escape from or approach the oyster, and:

[0033]

[0034] J(1) is the initial step length and jump value of each bitter fish; t max is the preset maximum number of iterations; U(t) is a random function, and:

[0035] U(t+1)=cos(t×cos -1(U(t)));

[0036] U(1) is equal to 1; the probability parameter P is:

[0037]

[0038] Where t is the number of iterations; a represents the restoration power; rand is a random number;

[0039] S2064, for each negative individual F j (-), iterative update is performed in the following way:

[0040]

[0041] Among them, [l,u] is the range of individual bitter fish values ​​for the initial generation of the bitter fish population, l and u are the lower and upper limits of the individual bitter fish values ​​respectively; M represents the average position of the fish school:

[0042]

[0043] S2065, each positive individual F i (+) The new bitter fish created is represented by:

[0044]

[0045] Positive individuals The new bitter fish created; R represents the distribution radius of the bitter fish individuals around the oyster shell, with a value between [0,2];

[0046] The probability of a newborn bitterling being hunted is:

[0047]

[0048] is the probability of a newborn bitterling being hunted, and f(·) is the objective function solved by the bitterling optimization algorithm;

[0049] Count the number of newborn bitter fish individuals K that survived after being hunted, and then use the K remaining surviving newborn bitter fish individuals Replacement away from oyster position K negative individuals Thus, a new bitter fish population of the t+1th iteration is formed;

[0050] S2066, set the number of iterations t=t+1, and then return to step S2062;

[0051] S2067, repeat the above steps S2062 to S2066 until the preset maximum number of iterations t=t is reached. max, the optimal solution of the bitter fish individual in the final iteration is obtained, that is, the bitter fish individual closest to the oyster position after the last iteration, as the optimized weight value.

[0052] As a preferred solution, in step S3, the objective function expression of the power distribution optimization model is:

[0053]

[0054] Among them, C totle is the total cost; C grid,i is the construction cost of the distribution network of the i-th grid; C install,i is the charging pile installation cost of the i-th grid; C trans,i is the power transmission loss cost of the i-th grid; M is the number of grids included in the distribution network grid map;

[0055] The construction cost of the distribution network of the i-th grid is C grid,i and power transmission loss cost C trans,i Calculate according to the following formula:

[0056] C grid,i =α1·L grid,i +α2·θ i ;

[0057] Among them, L grid,i represents the length of the distribution line of the i-th grid, θ i is the average slope angle of the i-th grid, α1 and α2 are coefficients used to represent the weight of distribution network construction and slope angle;

[0058] C trans,i =γ·P load,i ·L grid,i ;

[0059] Among them, P load,i represents the load of the ith grid, L grid,i represents the length of the distribution line of the ith grid, γ represents the proportionality coefficient of the power transmission loss;

[0060] The constraints of the power distribution optimization model include:

[0061] 1) Load balancing constraints:

[0062]

[0063] Among them, P load,i represents the load of the ith grid, P totle Represents the total load demand.

[0064] 2) Slope angle constraint:

[0065] θi ≥θ max ;

[0066] Among them, θ i is the average slope angle of the i-th grid, θ max is the maximum allowable slope angle in the planning area.

[0067] 3) Traffic flow constraints:

[0068] T grid,i ≥T min ;

[0069] Among them, T grid,i represents the minimum traffic flow in the i-th grid, T min Indicates the minimum traffic flow set in the planning area.

[0070] As a preferred solution, step S4 is specifically as follows:

[0071] S401. According to the fixed load distribution prediction results and the electric vehicle charging load prediction results, the grid areas in the distribution network grid optimization map are divided into load levels; the load levels include light load areas, medium load areas, uniform load areas, heavy load areas and overload areas, and the load level division is performed as follows:

[0072] The grid area is divided into light-load areas, the percentage of fixed load is in the range of [0%, 30%], and the percentage of electric vehicle charging load is in the range of [0%, 10%];

[0073] The grid area is divided into medium load area, the percentage of fixed load is in the range of (30%, 50%], and the percentage of electric vehicle charging load is in the range of (10%, 20%);

[0074] The grid area is divided into uniform load areas, the fixed load percentage range is (50%, 70%], and the electric vehicle charging load percentage range is (20%, 30%];

[0075] The grid area divided into heavy-load areas has a fixed load percentage range of (70%, 100%] and an electric vehicle charging load percentage range of (30%, 40%];

[0076] For the grid areas divided into overload areas, the percentage of fixed load is in the range of >100%, and the percentage of electric vehicle charging load is in the range of >40%.

[0077] S402. According to the load classification level standard, the grid of the grid division result is filled with grid color, the color of the overload area is filled with: R: 90%, G: 10%, B: 10%; the color of the heavy load area is filled with: R: 100%, G: 0%, B: 0%; the color of the even load area is filled with: R: 80%, G: 20%, B: 20%; the color of the medium load area is filled with: R: 70%, G: 30%, B: 30%; the color of the light load area is filled with: R: 100%, G: 30%, B: 0%;

[0078] S403. According to the above steps, the grid area load division result is obtained.

[0079] As a preferred solution, step S5 is specifically as follows:

[0080] S501. Based on the historical traffic flow data, the grid areas in the distribution network grid optimization map are divided into traffic thermal levels; the traffic thermal levels include completely unblocked areas, basically unblocked areas, normal operating areas, slow-moving areas, congested areas, and severely congested areas. The traffic thermal level division is performed as follows:

[0081] The grid area is divided into completely unblocked areas, and the traffic thermal value range is [0,50] vehicles / hour;

[0082] The grid area is divided into basically unblocked areas, and the traffic thermal value range is (50,200] vehicles / hour;

[0083] The grid area is divided into the normal operation area, and the traffic thermal value range is (200,500] vehicles / hour;

[0084] The grid area divided into slow-moving areas has a traffic heat value range of (500,1000] vehicles / hour;

[0085] The grid area divided into congested areas has a traffic thermal value range of (1000,2000] vehicles / hour;

[0086] For grid areas classified as heavily congested areas, the traffic thermal value range is >2000 vehicles / hour;

[0087] S502. According to the traffic thermal classification level standard, the grid division result is filled with grid colors. The chromaticity of the heavily congested area is: R: 43%, G: 0%, B: 100%; the chromaticity of the congested area is: R: 29%, G: 0%, B: 51%; the chromaticity of the slow-moving area is: R: 0%, G: 25%, B: 50%; the chromaticity of the normal driving area is: R: 0%, G: 50%, B: 100%; the chromaticity of the basically unblocked area is: R: 0%, G: 100%, B: 70%; the chromaticity of the completely unblocked area is: R: 0%, G: 40%, B: 60%;

[0088] S503. According to the above steps, the grid area traffic thermal division result is obtained.

[0089] As a preferred solution, step S6 is specifically as follows:

[0090] S601, according to the grid area load division result and the grid area traffic thermal division result, based on the division result of each grid area in the area to be planned, superimpose the grid area load division result corresponding to each grid area and the chromaticity of the grid area traffic thermal division result;

[0091] S602. Based on the color superposition results of each grid area, plan the location and capacity of the charging piles according to the following standards:

[0092] .

[0093] Compared with the prior art, the present invention has the following beneficial effects:

[0094] 1. The electric vehicle charging station planning method based on load and vehicle flow distribution of the present invention takes into account the superimposed influence of terrain data, historical load data and historical vehicle flow data in the planning, predicts the fixed load distribution of the charging station and the charging load of the electric vehicle with the historical load data, and combines the prediction results with the analysis of terrain data, adopts the distribution optimization model to optimize the distribution network grid map and then divide the grid area load to more accurately describe the influence of terrain data and historical load data on the grid planning layout of the charging station, and also combines the distribution network grid optimization map with the historical vehicle flow data to analyze the grid area traffic thermal distribution, and finally comprehensively considers the superimposed influence of the grid area load division and the grid area traffic thermal division results to determine the capacity planning scheme of the electric vehicle charging pile, so as to better ensure the accuracy and rationality of the electric vehicle charging station planning.

[0095] 2. The method of the present invention adopts a grid planning method based on the distribution network grid map, which fully considers the grid layout characteristics of charging stations in urban areas, and uses gridding to simplify the distribution information of charging stations in urban areas, which is more conducive to the use of distribution optimization models and grid area division superposition operations for effective calculation, thereby improving the effectiveness of the grid layout of charging stations in urban areas and the efficiency of charging station planning.

[0096] 3. In the method of the present invention, both the fixed load demand data and the electric vehicle charging load demand data are decomposed into multiple internal model function components, which are decomposed separately in combination with the ensemble empirical mode decomposition method and then predicted using a neural network, which helps to solve the problems of high complexity, difficulty in solving, and low prediction accuracy of electric vehicle charging station planning factors in urban areas with complex terrain. BRIEF DESCRIPTION OF THE DRAWINGS

[0097] In order to make the purpose, technical solution and advantages of the invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings, in which:

[0098] Figure 1 The present invention is a flow chart of a method for planning an electric vehicle charging station based on load and vehicle flow distribution. DETAILED DESCRIPTION

[0099] 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. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but only represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0100] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0101] In order to solve the problems of difficulty in solving the planning of electric vehicle charging stations in urban areas with complex terrain, long planning cycle, and insufficient planning rationality in the prior art, the present invention discloses a planning method for electric vehicle charging stations based on load and vehicle flow distribution, such as Figure 1 As shown, the following steps are included:

[0102] S1. Obtain terrain data, historical load data and historical vehicle flow data of the area to be planned; the terrain data includes slope information and a plane-based distribution network grid map; the historical load data includes fixed load demand data and electric vehicle charging load demand data.

[0103] In the specific implementation of the scheme of the present invention, the area to be planned can be an urban area in different terrain areas, whether it is an urban area in a plain area, a mountainous area, or a coastal area, it can be used as an area to be planned. Especially for urban areas with complex terrain such as mountainous areas and coastal areas, it is more suitable to use the method of the present invention to plan electric vehicle charging stations, because the method of the present invention takes into account terrain data including slope information and a plane-based distribution network grid map. In specific applications, terrain data can include altitude, slope angle, and geographic information system (GIS) data. The present invention focuses on using the slope angle data, and other terrain data can be used to correct the slope angle.

[0104] Different from plain areas, mountainous areas and coastal areas have a large number of gentle slopes, hills and other terrain factors, so that the installation of transmission lines is not carried out on a horizontal plane, and the line layout is restricted by the geographical environment. The laying of the line needs to bend, cross, tilt, etc. according to the distribution of buildings and changes in altitude. In the specific implementation of the scheme of the present invention, in order to obtain more accurate slope angle data, in this embodiment, spatial planning and special planning are combined to extract elements such as ridge lines, slope angles, ecological protection areas, administrative boundaries, and current medium-voltage distribution network lines, analyze spatial planning and special planning, and extract road networks, rivers, functional groups and other elements; due to the relatively high cost of medium-voltage distribution lines crossing mountains (ridges are too high, slopes are too large, etc.), it is not conducive to construction and maintenance, and the risk of natural disasters is also high. In the specific implementation of the scheme of the present invention, when the relative slope angle is greater than 30°, it is directly used as a grid boundary; and for the case where the slope angle is less than or equal to 30°, it is necessary to analyze the slope angle of the terrain and identify the steep slope area.

[0105] In the specific implementation of the scheme of the present invention, it is considered that in the actual distribution network scenario, the load can be divided into fixed load and dynamic load; wherein, the fixed load includes residential load and industrial load, and these load data are relatively stable; the dynamic load mainly refers to the dynamic load of electric vehicle charging at each charging station, and these dynamic loads are relatively random. Therefore, the present invention comprehensively considers the fixed load and dynamic load in the distribution network scenario, and needs to collect historical fixed load demand data and electric vehicle charging load demand data as historical load data to participate in the prediction, so as to obtain the fixed load distribution prediction results and electric vehicle charging load prediction results.

[0106] In the specific implementation of the solution of the present invention, the collected historical traffic flow data may include location information, speed and acceleration, remaining mileage of the vehicle, battery status and other data. In addition, other historical operating parameters of the electric vehicle may be selected as historical traffic flow data according to actual needs to participate in the necessary planning considerations.

[0107] S2. According to the historical load data, the fixed load and the electric vehicle charging load are predicted to obtain a fixed load distribution prediction result and an electric vehicle charging load prediction result.

[0108] In specific implementation, step S2 includes:

[0109] S201. Perform internal model function decomposition on the fixed load demand data and the electric vehicle charging load demand data in the historical load data to obtain fixed load internal model function components and electric vehicle charging load internal model function components respectively.

[0110] In the specific implementation of the scheme of the present invention, both the fixed load demand data and the electric vehicle charging load demand data can be decomposed into multiple internal model function components, and then predicted separately, thereby solving the problem of high complexity, difficulty in solving and low prediction accuracy of electric vehicle charging station planning factors in complex terrain urban areas.

[0111] S202, using the ensemble empirical mode decomposition method, respectively decompose each fixed load internal mode function component and the electric vehicle charging load internal mode function component to obtain a decomposed fixed load time series and an electric vehicle charging load time series.

[0112] This embodiment adopts Ensemble Empirical Mode Decomposition (EEMD), which can solve the mode mixing problem in the traditional empirical mode decomposition process. The decomposition principle of EEMD is: when the additional white noise is evenly distributed in the entire time-frequency space, the time-frequency space is composed of components of different scales divided by the filter group; when the signal is added with a uniformly distributed white noise background, the signal areas of different scales will be automatically mapped to the appropriate scale related to the background white noise; each independent test may produce very noisy results, because each component of the additional noise includes the signal and the additional white noise; since the noise is different in each independent test, when the overall mean of enough tests is used, the noise will be eliminated, and the overall mean will eventually be considered as the true result. With more and more tests, the additional noise is eliminated, and the only persistent and stable part is the signal itself. Therefore, the internal model function component of Gaussian white noise is directly added in the adaptive noise process of the ensemble empirical mode decomposition, which eliminates the reconstruction error in the traditional empirical mode decomposition process, and forms a signal analysis and processing method for fully integrated empirical mode decomposition adaptive noise.

[0113] S203. Determine the information entropy IE of each fixed load internal model function component and electric vehicle charging load internal model function component according to the decomposed fixed load time series and electric vehicle charging load time series, and reconstruct the fixed load extended internal model function component and electric vehicle charging load extended internal model function component with similar information entropy IE; the similar information entropy IE means that the difference between the information entropy IE of the two is within a preset difference range.

[0114] In a specific implementation, for any one of the fixed load time series and the electric vehicle charging load time series, the specific process of constructing an extended internal model function component with similar information entropy IE through step S203 is described as follows:

[0115] S2031, the white noise sequence W i Add it to the decomposed load time series to construct a new load time series, which is expressed as:

[0116] P (i) =P+β0E(W (i) ) (1)

[0117] Wherein, P is the load time series; E k (·) is the k-th order IMF component generated after the decomposition of the internal model function component; N(·) represents the local average value of the solved signal, W (i) represents a white noise sequence with a mean of 0; β is the standard deviation of the white noise;

[0118] S2032, using the ensemble empirical mode decomposition method to process the new load time series, to obtain a first group of residuals R1, expressed as:

[0119] R1=(N(P) (i) ) (2)

[0120] S2033, by subtracting the residual of the original data sequence, the first internal model function component d1 is obtained, which is expressed as:

[0121] d1=P-R1 (3)

[0122] S2034, according to the order of the internal mode components in the load time series, select the internal mode components of white noise and add them to the residuals obtained in step S2032 to obtain a second group of residuals R2, and use the ensemble empirical mode decomposition method to decompose R1+βE(W (i) ) is decomposed, where the second set of residuals R2 is expressed as:

[0123] R2=(N(R1+βE(W (i) ))) (4)

[0124] d2=k1-k2 (5)

[0125] Among them, d2 represents the second component of the internal mode component;

[0126] S2035, repeating steps S2031 to S2034 until the remaining residue R of the load time series is n Without further decomposition, the new load time series after decomposition is obtained, which is expressed as:

[0127]

[0128] S2036, determining the IE of each IMF component according to the new load time series to reconstruct the IMF components with similar IE, the calculation formula is as follows:

[0129]

[0130] Wherein, IE is the information entropy value, X is the load data set; N is the number of the loaded load time series data; P k is the percentage of the kth load data in the data set of the load time series.

[0131] By reconstructing the fixed load extended internal model function components and the electric vehicle charging load extended internal model function components with similar information entropy IE, the basic data used for fixed load distribution prediction and electric vehicle charging load prediction can be reasonably expanded, the data prediction granularity can be refined, and the prediction accuracy can be improved.

[0132] S204. Use the information entropy IE of the fixed load internal model function component and the fixed load extended internal model function component as the input of the preset LSTM model to predict, and obtain the first fixed load distribution prediction result; use the information entropy IE of the electric vehicle charging load internal model function component and the electric vehicle charging load extended internal model function component as the input of the LSTM model to predict, and obtain the first electric vehicle charging load prediction result.

[0133] LSTM neural network (Long Short-Term Memory) is a special type of recurrent neural network (RNN) proposed by Hochreiter and Schmidhuber in 1997. It aims to solve the problem of gradient vanishing or gradient exploding encountered by traditional RNN when processing long sequence data. The core of LSTM network is the mechanism of three gates: forget gate, input gate, and output gate. These gates control the flow of information in an adaptive way, so as to capture long-term dependent information. Among them, the forget gate is used to forget some information output by the previous unit that affects the load forecasting results; the input gate determines the information stored in the unit with the new input information; the storage unit is used to store useful information transmitted from the previous LSTM unit to the unit; the output gate is used to output the processed information to the next LSTM unit. LSTM network continuously optimizes the weights in the unit during training and shares the weights among all units. Therefore, when the model parameters are fixed, the integral scale changes continuously at different times, which solves the problem of gradient vanishing and reduces the frequency of gradient explosion.

[0134] Taking IE as input, the expression calculated by the LSTM model is as follows:

[0135]

[0136] In the formula, i t is the input gate; x t is the input at the current moment, which is the IE of each IMF component; σ is the activation function of tanh, and the value range of σ is [-1,1]; W xs , W xi , W xf , W xo Both represent input X t The weight matrix of t is the output at the current moment; H t-1 is the output of the previous moment; H xs , H xi , H xf , H xo Both represent Ht-1 The weight matrix of i 、b f 、b o 、b s are all deviation vectors; f t is the forget gate, s t Represents the input unit vector at the current moment.

[0137] S205. Use the information entropy IE of the fixed load internal model function component and the fixed load extended internal model function component as the input of the preset GMDH model to predict and obtain the second fixed load distribution prediction result; use the information entropy IE of the electric vehicle charging load internal model function component and the electric vehicle charging load extended internal model function component as the input of the GMDH model to predict and obtain the second electric vehicle charging load prediction result.

[0138] The GMDH model (Group Method of Data Handling) is a multi-layer feedforward neural network (the number of network layers can reach 8 layers). It uses polynomial iteration to obtain the nonlinear relationship between input and output, so it is also called a polynomial network. The GMDH neural network is layered, and its initialization does not require setting the number of "neurons" and the number of network layers. It uses adaptive linear functions (polynomial functions) as neurons. Its training process is a process of continuously generating active neurons. The generated active neurons are screened and combined to generate neurons in the next layer until an optimal prediction model is obtained. The neurons in the input layer of the network are only responsible for transmitting input signals to the neurons in the middle layer; each neuron in the hidden layer and each neuron in the output layer are only connected to two neurons in the previous layer. Therefore, each layer of neurons in the GDMD neural network is independent.

[0139] Taking IE as input, the expression calculated by the GDMH model is as follows:

[0140] The input data is defined as P = (P1, P2, P3, ..., P N ), the predicted output is The actual output is O i The relationship between the neurons of GDMH is shown in the following formula:

[0141]

[0142] In the formula, O k,l represents the output of the lth neuron in the kth layer of the network; O k-1,i represents the output of the i-th neuron in the k-th layer of the network; O k-1,j represents the output of the jth neuron in the kth layer of the network; O k-1i Indicates input (P i), is the IE of each IMF component, a k,1 、b k,1 、c k,1 ,d k,1 、e k,1 are polynomial coefficients.

[0143] The quadratic polynomial O of the GMDH model is expressed as:

[0144] O=Pδ (10)

[0145]

[0146] Among them, σ is the polynomial coefficient of the binomial, and P is the matrix of the load time series;

[0147] In order to minimize the error between the predicted output and the actual output, the GMDH neural network obtains the coefficients of the quadratic polynomial in the above formula through regression analysis, and optimizes the polynomial coefficients of the binomial using the least squares method. The optimization formula is:

[0148]

[0149] in, is the expected output, O i is the actual prediction output; R represents the prediction error. When R is the smallest, the optimal quadratic polynomial is obtained, where the polynomial coefficient β of the binomial is calculated using the least squares formula, and the calculation formula is:

[0150] σπ=(P T P) -1 P T O (13)

[0151] Among them, P T is the transpose of the matrix of the load time series.

[0152] In the neural network of the GMDH model, any two neurons of a neural layer are used as input to generate neurons of a new neural layer; if the number of neurons m in the current neural layer is less than or equal to the size M of the preset elite pool, the neurons of the current neural layer are not processed; if the number of neurons m in the current neural layer is greater than the size M of the preset elite pool, the mean square error D of the neurons in the next layer is calculated. i , the calculation formula is as follows:

[0153]

[0154] Among them, C N represents the total number of neural layers, and the GMDH model is used to calculate the mean square error D according to the output iThe size of each neuron is sorted in ascending order, and according to the preset elite pool size M of the set, the first M neurons generated in the current neural layer are selected as input neurons of the next layer, and other neurons except the M neurons in the current neural layer are discarded to simplify the structure of the neural network.

[0155] S206. Perform a weighted summation on the first fixed load distribution prediction result and the second fixed load distribution prediction result to obtain a final fixed load distribution prediction result; perform a weighted summation on the first electric vehicle charging load prediction result and the second electric vehicle charging load prediction result to obtain a final electric vehicle charging load prediction result.

[0156] In a specific implementation, step S206 is specifically as follows: firstly, the first weight value of the weighted sum of the first fixed load distribution prediction result and the second fixed load distribution prediction result, and the second weight value of the weighted sum of the first electric vehicle charging load prediction result and the second electric vehicle charging load prediction result are optimized using the bitter fish optimization algorithm, and then the optimized first weight value and second weight value are used to weightedly sum and calculate the fixed load distribution prediction result and the electric vehicle charging load prediction result.

[0157] The Bitterling Fish Optimization (BFO, see the literature "Zareian L, Rahebi y, Shayegan M J. Bitterling fish optimization (BFO) algorithmically. Multimedia Tools and Applicaions, 2024: 1-34") is a swarm intelligence optimization algorithm inspired by nature. Bitter fish are a typical example of smart behavior in nature for survival. The algorithm models the Bitter Fish Optimization algorithm based on the mating behavior of these fish. Experiments and implementations on various benchmark functions show that the Bitter Fish Optimization algorithm is more accurate and has better optimization effects than commonly used optimization algorithms such as the Gray Wolf Optimization Algorithm Q, the Whale Optimization Algorithm, the Butterfly Optimization Algorithm, and the Harris Hawk Optimization Algorithm.

[0158] The BFO algorithm is inspired by the unique reproduction mechanism of bitterling fish. This fish feeds on oysters, and it is the males' responsibility to find oyster shells that are suitable for spawning. They need to find larger oysters with more space to accommodate their eggs. When these fish find one or more oysters, they will fight back against their enemies. The reason for the aggressive behavior of male species is that some other male species also tend to take the oysters for themselves. In fighting mode, the male fish's body color darkens and defends its territory with all its strength. During the mating stage, a phenomenon called sexual selection occurs. During the sexual selection process, female fish usually choose male fish with larger body size and more beautiful body color. At this time, the female fish lays eggs in the oysters, and then the male fish fertilizes the eggs with sperm to produce new small fish.

[0159] Specifically, in step S206, for any one of the first weight value and the second weight value, the specific method of using the bitter fish optimization algorithm for optimization is:

[0160] S2061, initializing the population;

[0161] Randomly generate n initial weight values ​​as the bitter fish individuals of the bitter fish population, i = 1, 2, ..., n, and get the initialized bitter fish population Pop, that is:

[0162] Pop=[F i |i=1,2,…,n],F i =[f i,k |k=1,2,…,m];(15)

[0163] Among them, F i ∈Pop is any i-th bitter fish individual in the bitter fish population, and the individuals in the bitter fish population are the weight values ​​to be optimized; f i,k is any i-th bitter fish individual F in the bitter fish population i The value of the kth dimension, each dimension value represents a decision variable; n is the number of bitter fish individuals contained in the bitter fish population in the bitter fish optimization algorithm, and m is the dimension of each bitter fish individual in the bitter fish population;

[0164] Then, the number of iterations t=0 is initialized, and the process goes to step S2062 to start the iteration of the bitter fish population.

[0165] S2062, classification of individual behaviors of populations;

[0166] Randomly divide half of the bitter fish individuals in the bitter fish population into positive individuals and mark them as F. i (+), and the other half of the bitter fish individuals are negative individuals and are marked as F j (-). Positive individual F i (+)∈Pop are those bitterlings searching for and trying to catch oysters; negative individuals F j(-)∈Pop are those bitterlings that escaped or failed to capture oysters.

[0167] S2063, search for and capture oysters;

[0168] Any male bittern is eligible to be located in an area with more oyster shells; bittern individuals can roam in search of oysters, in which case the positive individual F i (+) Targeting an oyster and moving toward it; in this case, the target oyster is not noticed by the other fish and is moved toward it by the fish F i (+) Capture; Oysters are captured by positive individual F i (+) The captured occupied state is expressed by the following formula (16):

[0169]

[0170] In the formula, and They are positive individuals F i (+) Positions in the tth and t+1th iterations; is a randomly selected oyster position; is the optimal solution of the bittern individual in the tth iteration, that is, the bittern individual closest to the oyster position in the tth iteration; δ and r are random numbers between 0 and 1. J(t) is the number of steps or speed that the bittern moves to escape or approach the oyster. J(t) will decrease as the number of algorithm iterations t increases. This is because after a period of time, the activity of male fish will decrease because they have successfully mated. As the number of iterations goes by, reducing J will reduce the global search to a local search. The reduction parameter J is expressed by formula (17):

[0171]

[0172] J(1) is the initial step length and jump value of each bitter fish. In the specific implementation, J(t)=1; t max is the preset maximum number of iterations; J(t) is the number of steps each fish takes to find oysters in repetition t. The random function U(t) is used to generate a random sequence, expressed as (18):

[0173] U(t+1)=cos(t×cos -1 (U(t))) (18)

[0174] U(1) can be considered equal to 1; in summary, (17) and (18) can be expressed by (19):

[0175]

[0176] In the bitter fish optimization algorithm, the optimal solution is repeatedly searched over time and in more space around the optimal solution. Therefore, it is necessary to reduce the probability parameter P over time to the second criterion of formula (16), under the condition that r>P. At this time, the probability parameter P can be expressed by formula (20):

[0177]

[0178] Wherein, t is the number of iterations; a represents the restoration power, and in a specific implementation, a=0.9 can be taken; and rand is a random number.

[0179] S2064, escape behavior;

[0180] For negative individuals F j (-) to perform mathematical simulation of escape, formula (21) can be used to describe the individual bittern that escapes or fails to capture oysters:

[0181]

[0182] Among them, [l,u] is the range of individual bitter fish values ​​for the initial generation of the bitter fish population, and l and u are the lower and upper limits of the individual bitter fish values, respectively.

[0183] M represents the average position of the fish school; that is, after the oysters escape, the bittern can only look for oysters in the average space, which can be expressed as a random optimal space search problem, expressed by formula (22):

[0184]

[0185] S2065, spawning behavior;

[0186] After finding a suitable oyster, the male fish can attract the female fish to lay eggs in the oyster. The male fish fertilizes the eggs, which creates new fish around them. This process is expressed by formula (23):

[0187]

[0188] Positive individuals The new bitter fish created; R represents the distribution radius of the bitter fish individuals around the oyster shell, and its value is between [0,2]. Usually, the initial value is R=2, and it decreases as the number of iterations increases.

[0189] Hunting behavior;

[0190] At the same time, an oyster can hold several small fish at the same time, but weaker fish will be hunted by the oyster or predators. The probability of losing a small fish is:

[0191]

[0192] is the probability of a newborn bitterling being hunted, and f(·) is the objective function solved by the bitterling optimization algorithm;

[0193] Count the number of newborn bitter fish individuals K that survived after being hunted, and then use the K remaining surviving newborn bitter fish individuals Replacement away from oyster position K negative individuals Thus, a new bitter fish population of the t+1th iteration is formed.

[0194] S2066, Iteration:

[0195] Let the number of iterations t=t+1, and then return to step S2062; in this way, the bitter fish population enters the next iteration.

[0196] S2067, Convergence:

[0197] Repeat the above steps S2062 to S2066 until the preset maximum number of iterations t=t is reached. max , the optimal solution of the bitter fish individual in the final iteration is obtained, that is, the bitter fish individual closest to the oyster position after the last iteration, as the optimized weight value.

[0198] The present invention adopts the bitter fish optimization algorithm to optimize the first weight value and the second weight value, and can find the optimal weight value, so that the fixed load distribution prediction results and the electric vehicle charging load prediction results obtained by weighted sum calculation are more accurate, which is beneficial to the accurate optimization of the distribution network grid optimization map based on the fixed load distribution prediction results and the electric vehicle charging load prediction results in the subsequent process.

[0199] S3. Based on the slope information, the distribution network grid map, and the obtained fixed load distribution prediction results and electric vehicle charging load prediction results, a preset distribution optimization model is used to optimize the distribution network grid optimization map with total cost minimization as the objective function.

[0200] In the specific implementation of the solution of the present invention, a grid area in the distribution network grid map is regarded as a charging station planning area (no matter how many real charging stations there are in the actual area, they are regarded as the same grid area to implement unified charging station planning), and the total cost is minimized as the goal, and the preset distribution optimization model is used for optimization. Assuming that the number of grids contained in the distribution network grid map is, the objective function expression of the distribution optimization model is:

[0201]

[0202] Among them, C totle is the total cost; Cgrid,i is the construction cost of the distribution network of the i-th grid; C install,i is the charging pile installation cost of the i-th grid; C trans,i is the power transmission loss cost of the i-th grid; M is the number of grids included in the distribution network grid map.

[0203] Cost function definition:

[0204] The cost of each grid is defined as follows:

[0205] The construction cost of the distribution network of the i-th grid is C grid,i and power transmission loss cost C trans,i Calculate according to the following formula:

[0206] C grid,i =α1·L grid,i +α2·θ i ;

[0207] Among them, L grid,i represents the length of the distribution line of the i-th grid, θ i is the average slope angle of the i-th grid, α1 and α2 are coefficients used to represent the weight of distribution network construction and slope angle;

[0208] C trans,i =γ·P load,i ·L grid,i ;

[0209] Among them, P load,i represents the load of the ith grid, L grid,i represents the length of the distribution line of the ith grid, γ represents the proportionality coefficient of the power transmission loss;

[0210] The charging pile installation cost C of the i-th grid install,i , then you can calculate based on the actual installation and construction cost of the charging pile.

[0211] Constraints of the power distribution optimization model:

[0212] In order to ensure the rationality of mesh division, the following constraints need to be considered:

[0213] 1) Load balancing constraints

[0214]

[0215] Among them, P load,i represents the load of the ith grid, P totle Represents the total load demand.

[0216] 2) Slope angle constraint

[0217] θ i≥θ max ;

[0218] Among them, θ i is the average slope angle of the i-th grid, θ max is the maximum allowable slope angle in the planning area.

[0219] 3) Traffic flow constraints

[0220] T grid,i ≥T min ;

[0221] Among them, T grid,i represents the minimum traffic flow in the i-th grid, T min Indicates the minimum traffic flow set in the planning area.

[0222] Through the optimization processing of the above-mentioned distribution optimization model, an optimized distribution network grid optimization map can be obtained, so that the load level and traffic heat division of each grid area can be carried out according to the distribution network grid optimization map to determine the capacity layout of electric vehicle charging piles in each grid area, so as to obtain the final planning scheme.

[0223] The grid planning method based on the distribution network grid map not only fully considers the grid layout characteristics of charging stations in urban areas, but also uses gridding to simplify the distribution information of charging stations in urban areas. It is more conducive to effective calculation using distribution optimization models and grid area division superposition operations, thereby improving the effectiveness of the grid layout of charging stations in urban areas and the efficiency of charging station planning.

[0224] S4. According to the fixed load distribution prediction result and the electric vehicle charging load prediction result, the grid area in the distribution network grid optimization map is divided into load levels to obtain a grid area load division result.

[0225] In specific implementation, step S4 is specifically as follows:

[0226] S401. According to the fixed load distribution prediction results and the electric vehicle charging load prediction results, the grid areas in the distribution network grid optimization map are divided into load levels; the load levels include light load areas, medium load areas, uniform load areas, heavy load areas and overload areas, and the load level division is performed as follows:

[0227] For the grid area divided into light-load areas, the percentage of fixed load is in the range of [0%, 30%], and the percentage of electric vehicle charging load is in the range of [0%, 10%]; this means that the load in the grid area is far lower than the planned capacity, and most of them are areas with low electricity demand (such as suburbs and industrial areas during non-peak hours);

[0228] For the grid area divided into medium load area, the percentage of fixed load is in the range of (30%, 50%), and the percentage of electric vehicle charging load is in the range of (10%, 20%); this means that the load level in the grid area is moderate and the grid load is in the normal working range, such as ordinary residential areas or light industrial areas;

[0229] The grid area divided into uniform load areas has a fixed load percentage range of (50%, 70%], and an electric vehicle charging load percentage range of (20%, 30%]; this means that the load in the grid area is close to a balanced operating state, which is a good utilization range of the planned capacity, and the system security is relatively high;

[0230] For the grid area divided into heavy-load areas, the percentage of fixed load is in the range of (70%, 100%], and the percentage of electric vehicle charging load is in the range of (30%, 40%]; this means that the load in the grid area is close to the upper limit of the planned capacity, which may be a high-density residential area or a commercial concentrated area;

[0231] Grid areas classified as overloaded areas have a fixed load percentage range of >100% and an EV charging load percentage range of >40%; this represents a grid area where the load exceeds the planned capacity, usually due to short-term peaks or excessive EV charging.

[0232] S402. According to the load classification level standard, the grid of the grid division result is filled with a grid color, and the chromaticity of the overload area is: R: 90%, G: 10%, B: 10% (dark red); the chromaticity of the heavy load area is: R: 100%, G: 0%, B: 0% (red); the chromaticity of the even load area is: R: 80%, G: 20%, B: 20% (medium red); the chromaticity of the medium load area is: R: 70%, G: 30%, B: 30% (light red); the chromaticity of the light load area is: R: 100%, G: 30%, B: 0% (orange);

[0233] S403. According to the above steps, the grid area load division result is obtained.

[0234] S5. Based on the historical vehicle flow data, the grid areas in the distribution network grid optimization map are divided into traffic thermal levels to obtain grid area traffic thermal division results.

[0235] In specific implementation, step S5 is specifically as follows:

[0236] S501. Based on the historical traffic flow data, the grid areas in the distribution network grid optimization map are divided into traffic thermal levels; the traffic thermal levels include completely unblocked areas, basically unblocked areas, normal operating areas, slow-moving areas, congested areas, and severely congested areas. The traffic thermal level division is performed as follows:

[0237] The grid area divided into completely unblocked areas has a traffic thermal value range of [0,50] vehicles / hour; this means that the traffic volume in the grid area is extremely low, there is almost no traffic congestion, and vehicles can travel freely;

[0238] The grid area divided into the basically unblocked area has a traffic thermal value range of (50,200] vehicles / hour; this means that the traffic volume in the grid area is low, the vehicles travel smoothly, and there are occasional slight speed fluctuations;

[0239] The grid area divided into normal operation areas has a traffic thermal value range of (200,500] vehicles / hour; this means that the traffic volume in the grid area is at a moderate level, the traffic is basically smooth, and some sections may experience a slight reduction in speed;

[0240] The grid area divided into slow-moving areas has a traffic heat value range of (500,1000] vehicles / hour; this means that the traffic volume in the grid area is high, and traffic on some sections of the road begins to slow down, with small-scale queues.

[0241] The grid area divided into congested areas has a traffic heat value range of (1000,2000] vehicles / hour; this means that the traffic flow in the grid area is close to road saturation, vehicles are moving slowly, and traffic pressure is high on most sections of the road;

[0242] The grid area classified as a severely congested area has a traffic thermal value range of >2000 vehicles / hour; this means that the traffic volume in the grid area seriously exceeds the road carrying capacity, vehicles are almost stopped, and traffic is paralyzed;

[0243] S502. According to the traffic thermal classification level standard, the grid division result is filled with grid color, and the chromaticity of filling the severely congested area is: R: 43%, G: 0%, B: 100% (cyan blue); the chromaticity of filling the congested area is: R: 29%, G: 0%, B: 51% (cyan); the chromaticity of filling the slow-moving area is: R: 0%, G: 25%, B: 50% (glazed color); the chromaticity of filling the normal driving area is: R: 0%, G: 50%, B: 100% (light blue); the chromaticity of filling the basically unblocked area is: R: 0%, G: 100%, B: 70% (cyan); the chromaticity of filling the completely unblocked area is: R: 0%, G: 40%, B: 60% (empty color);

[0244] S503. According to the above steps, the grid area traffic thermal division result is obtained.

[0245] S6. According to the grid area load division result and the grid area traffic thermal division result, a capacity planning scheme for electric vehicle charging piles in different grid areas in the area to be planned is obtained.

[0246] In the specific implementation of the scheme of the present invention, since the above step S4 has obtained the grid area load division result, step S5 has obtained the grid area traffic heat division result, and the load level division and traffic heat division of different grid areas are distinguished by color. The heavier the load and the higher the traffic heat, the darker the color is used for marking. In this step, the color of the load level division and the color of the traffic heat division can be further combined and superimposed on each other to intuitively reflect the layout of the electric vehicle charging pile capacity in each grid area, so as to obtain the final planning scheme. Therefore, in this step S6, the electric vehicle charging pile capacity planning of different grid areas in the area to be planned is implemented in the following way:

[0247] S601, according to the grid area load division result and the grid area traffic thermal division result, based on the division result of each grid area in the area to be planned, superimpose the grid area load division result corresponding to each grid area and the chromaticity of the grid area traffic thermal division result;

[0248] S602. Based on the color superposition results of each grid area, plan the location and capacity of the charging piles according to the following standards:

[0249]

[0250] The power-traffic heat map of each grid area obtained by chromaticity superposition comprehensively considers the superposition of grid area load division results and grid area traffic heat division results. The electric vehicle charging pile capacity planning scheme of each grid area obtained by this has the following technical advantages:

[0251] 1. Dynamically match power load and traffic demand. Take power load and traffic flow into consideration to ensure that charging piles are reasonably distributed within the carrying capacity of the power system. Reduce or limit the planning of charging piles in overloaded or heavy-loaded areas, and increase the deployment of charging piles in evenly loaded or lightly loaded areas to avoid further aggravating the pressure on the power grid. Avoid overloading the power grid due to improper planning of charging piles, while meeting the charging needs of vehicles in high traffic heat areas.

[0252] 2. Clear regional classification and targeted planning. Through clear zoning strategies (light load, medium load, uniform load, heavy load, overload) and corresponding traffic thermal levels, the capacity and functional distribution of charging piles are more accurate. For example:

[0253] Light-load area: The power pressure is low and the traffic demand is average. 100% of the charging pile capacity can be planned, and it can even be used as a key area for the promotion of electric vehicle charging.

[0254] Medium-load area: Traffic heat is moderate, power load is stable, and charging piles are deployed at 80% of the planned capacity to ensure a balance between supply and demand.

[0255] Uniform load area: The load is close to the design capacity, which is suitable for the planned deployment of 60% to 80% of charging piles, while taking into account the stability of the power grid.

[0256] Heavy load and overload areas: The grid is under great pressure, and load transfer is guided by limiting the construction of charging piles or only discharging mode.

[0257] 3. Make full use of the resources in evenly loaded and lightly loaded areas. Plan a higher proportion of charging piles (up to 100%) in evenly loaded and lightly loaded areas to avoid idle power resources, and at the same time reduce the power pressure in other areas by guiding traffic to these areas. Improve the utilization rate of the power grid, optimize the global layout of charging piles, and alleviate the charging pressure in heavy or overloaded areas.

[0258] 4. Consider both traffic flow and service convenience. In areas with high traffic flow (such as heavily congested areas), even if the grid load is high, the convenience of service can be improved by properly planning the discharge mode or a small number of charging piles. This ensures the efficiency of urban traffic operation and prevents traffic from further deteriorating due to insufficient charging piles.

[0259] 5. Alleviate urban congestion and improve resource allocation efficiency. Reduce the planning of charging piles or only provide discharge functions in heavily congested areas (high traffic heat value), reduce vehicle dwelling time, and induce vehicles to charge in evenly loaded or lightly loaded areas. Optimize urban traffic flow distribution and reduce the dual pressure of congestion and power load.

[0260] S7. According to the obtained electric vehicle charging pile capacity planning scheme, the capacity of the electric vehicle charging piles distributed in different grid areas in the planning area is planned.

[0261] In the specific implementation of the solution of the present invention, based on the obtained electric vehicle charging pile capacity planning solution, the corresponding electric vehicle charging pile capacity planning strategy is implemented for the charging station in each grid area, so as to achieve reasonable planning of the charging pile capacity in all charging stations in the planning area.

[0262] The electric vehicle charging station planning method of the present invention is based on load and vehicle flow distribution. The superposition influence of terrain data, historical load data and historical vehicle flow data is considered in the planning. The fixed load distribution of the charging station and the charging load of the electric vehicle are predicted by the historical load data. The prediction result is combined with the analysis of terrain data. The distribution network grid map is optimized by using the distribution optimization model and then the grid area load is divided to more accurately describe the influence of terrain data and historical load data on the grid planning layout of the charging station. At the same time, the grid area traffic heat distribution is analyzed in combination with the distribution network grid optimization map and the historical vehicle flow data. Finally, the superposition influence of the grid area load division and the grid area traffic heat division results is comprehensively considered to determine the capacity planning scheme of the electric vehicle charging pile, so as to better ensure the accuracy and rationality of the electric vehicle charging station planning.

[0263] A person skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the relevant hardware through a program, and the program can be stored in a computer-readable storage medium, which can include: ROM, RAM, disk or CD, etc.

[0264] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit the technical solution. Those skilled in the art should understand that those modifications or equivalent substitutions of the technical solution of the present invention that do not depart from the purpose and scope of the technical solution should be included in the scope of the claims of the present invention.

Claims

1. A method for planning electric vehicle charging stations based on load and traffic flow distribution, characterized in that: The steps include: S1. Obtain terrain data, historical load data and historical vehicle flow data of the area to be planned; the terrain data includes slope information and a plane-based distribution network grid map; the historical load data includes fixed load demand data and electric vehicle charging load demand data; S2. Predicting the fixed load and the electric vehicle charging load according to the historical load data to obtain a fixed load distribution prediction result and an electric vehicle charging load prediction result; S3, according to the slope information, the distribution network grid map, and the obtained fixed load distribution prediction results and electric vehicle charging load prediction results, a preset distribution optimization model is used to optimize the distribution network grid optimization map with total cost minimization as the objective function; S4. According to the fixed load distribution prediction result and the electric vehicle charging load prediction result, the grid area in the distribution network grid optimization map is divided into load levels to obtain a grid area load division result; S5. Based on the historical vehicle flow data, the grid areas in the distribution network grid optimization map are divided into traffic thermal levels to obtain grid area traffic thermal division results; S6. Obtaining a capacity planning scheme for electric vehicle charging piles in different grid areas in the area to be planned based on the grid area load division result and the grid area traffic thermal division result; S7. According to the obtained electric vehicle charging pile capacity planning scheme, the capacity of the electric vehicle charging piles distributed in different grid areas in the planning area is planned.

2. The electric vehicle charging station planning method based on load and traffic flow distribution according to claim 1 is characterized in that: Step S2 is specifically as follows: S201, performing internal model function decomposition on the fixed load demand data and the electric vehicle charging load demand data in the historical load data, respectively, to obtain fixed load internal model function components and electric vehicle charging load internal model function components; S202, using the ensemble empirical mode decomposition method to decompose each fixed load internal mode function component and the electric vehicle charging load internal mode function component, respectively, to obtain a decomposed fixed load time series and an electric vehicle charging load time series; S203, according to the decomposed fixed load time series and electric vehicle charging load time series, determine the information entropy IE of each fixed load internal model function component and electric vehicle charging load internal model function component, and reconstruct the fixed load extended internal model function component and electric vehicle charging load extended internal model function component with similar information entropy IE; the similar information entropy IE means that the difference between the information entropy IE of the two is within a preset difference range; S204, using the information entropy IE of the fixed load internal model function component and the fixed load extended internal model function component as the input of the preset LSTM model to predict, and obtain the first fixed load distribution prediction result; using the information entropy IE of the electric vehicle charging load internal model function component and the electric vehicle charging load extended internal model function component as the input of the LSTM model to predict, and obtain the first electric vehicle charging load prediction result; S205, using the information entropy IE of the fixed load internal model function component and the fixed load extended internal model function component as the input of the preset GMDH model to predict, and obtain a second fixed load distribution prediction result; using the information entropy IE of the electric vehicle charging load internal model function component and the electric vehicle charging load extended internal model function component as the input of the GMDH model to predict, and obtain a second electric vehicle charging load prediction result; S206. Perform a weighted summation on the first fixed load distribution prediction result and the second fixed load distribution prediction result to obtain a final fixed load distribution prediction result; perform a weighted summation on the first electric vehicle charging load prediction result and the second electric vehicle charging load prediction result to obtain a final electric vehicle charging load prediction result.

3. The electric vehicle charging station planning method based on load and traffic flow distribution according to claim 1 is characterized in that: Step S206 is specifically as follows: First, the first weight value of the weighted sum of the first fixed load distribution prediction result and the second fixed load distribution prediction result, and the second weight value of the weighted sum of the first electric vehicle charging load prediction result and the second electric vehicle charging load prediction result are optimized using the bitter fish optimization algorithm, and then the optimized first weight value and second weight value are used to weighted sum and calculate the fixed load distribution prediction result and the electric vehicle charging load prediction result.

4. The electric vehicle charging station planning method based on load and vehicle flow distribution according to claim 3 is characterized in that: In step S206, for any one of the first weight value and the second weight value, the specific method of using the bitter fish optimization algorithm for optimization is: S2061, randomly generate n initial weight values ​​as bitter fish individuals of the bitter fish population, i = 1, 2, ..., n, and obtain the initialized bitter fish population Pop, that is: Pop=[F i |i=1,2,…,n],F i =[f i,k |k=1,2,…,m]; Among them, F i ∈Pop is any i-th bitter fish individual in the bitter fish population, and the individuals in the bitter fish population are the weight values ​​to be optimized; f i,k is any i-th bitter fish individual F in the bitter fish population i The value in the kth dimension; n is the number of bitter fish individuals contained in the bitter fish population in the bitter fish optimization algorithm, and m is the dimension of each bitter fish individual in the bitter fish population; Then, the number of iterations is initialized to t=1, and the process goes to step S2062; S2062. Randomly divide half of the bitter fish individuals in the bitter fish population into positive individuals and mark them as F. i (+), and the other half of the bitter fish individuals are negative individuals and are marked as F j (-); S2063, for each positive individual F i (+), iterative update is performed in the following way: In the formula, F i t (+) and F i t+1 (+) are positive individuals F i (+) Position in the tth and t+1th iterations; F i + is a randomly selected oyster position; F i * is the optimal solution of the bittern in the tth iteration, that is, the bittern closest to the oyster in the tth iteration; δ and r are random numbers between 0 and 1; J(t) is the number of steps or speed that the bittern moves to escape from or approach the oyster, and: J(1) is the initial step length and jump value of each bitter fish; t max is the preset maximum number of iterations; U(t) is a random function, and: U(t+1)=cos(t×cos -1 (U(t))); U(1) is equal to 1; the probability parameter P is: Where t is the number of iterations; a represents the restoration power; rand is a random number; S2064, for each negative individual F j (-), iterative update is performed in the following way: Among them, [l,u] is the range of individual bitter fish values ​​for the initial generation of the bitter fish population, l and u are the lower and upper limits of the individual bitter fish values ​​respectively; M represents the average position of the fish school: S2065, each positive individual F i (+) The new bitter fish created is represented by: F i t+1 (new)=F i (+)+R×rand(0,1); F i t+1 (new) indicates the positive individual F i t (+) The number of new bitter fish created; R represents the distribution radius of bitter fish individuals around the oyster shell, with a value between [0,2]; The probability of a newborn bitterling being hunted is: d(F i t+1 (new)) is the probability of a new bitterling being hunted, and f(·) is the objective function solved by the bitterling optimization algorithm; Count the number of newborn bitter fish individuals K that survive after being hunted, and then use the K remaining surviving newborn bitter fish individuals F i t +1 (new) Replace the far oyster position F i t+ K negative individuals F i t+1 (-) Thus, a new bitter fish population of the t+1th iteration is formed; S2066, set the number of iterations t=t+1, and then return to step S2062; S2067, repeat the above steps S2062 to S2066 until the preset maximum number of iterations t=t is reached. max , the optimal solution of the bitter fish individual in the final iteration is obtained, that is, the bitter fish individual closest to the oyster position after the last iteration, as the optimized weight value.

5. The electric vehicle charging station planning method based on load and vehicle flow distribution according to claim 4 is characterized in that: In step S3, the objective function expression of the power distribution optimization model is: Among them, C totle is the total cost; C grid,i is the construction cost of the distribution network of the i-th grid; C install,i is the charging pile installation cost of the i-th grid; C trans,i is the power transmission loss cost of the i-th grid; M is the number of grids included in the distribution network grid map; The construction cost of the distribution network of the i-th grid is C grid,i and power transmission loss cost C trans,i Calculate according to the following formula: C grid,i =α1·L grid,i +α2·θ i ; Among them, L grid,i represents the length of the distribution line of the i-th grid, θ i is the average slope angle of the i-th grid, α1 and α2 are coefficients used to represent the weight of distribution network construction and slope angle; C trans,i =γ·P load,i ·L grid,i ; Among them, P load,i represents the load of the ith grid, L grid,i represents the length of the distribution line of the ith grid, γ represents the proportionality coefficient of the power transmission loss; The constraints of the power distribution optimization model include: 1) Load balancing constraints: Among them, P load,i represents the load of the ith grid, P totle Represents the total load demand. 2) Slope angle constraint: i i ≥θ max ; Among them, θ i is the average slope angle of the i-th grid, θ max is the maximum allowable slope angle in the planning area. 3) Traffic flow constraints: T grid,i ≥T min ; Among them, T grid,i represents the minimum traffic flow in the i-th grid, T min Indicates the minimum traffic flow set in the planning area.

6. The electric vehicle charging station planning method based on load and vehicle flow distribution according to claim 1 is characterized in that: Step S4 is specifically as follows: S401, based on the fixed load distribution prediction result and the electric vehicle charging load prediction result, the grid area in the distribution network grid optimization map is divided into load levels; the load levels include light load area, medium load area, uniform load area, heavy load area and overload area, and the load level division is performed as follows: The grid area is divided into light-load areas, the percentage of fixed load is in the range of [0%, 30%], and the percentage of electric vehicle charging load is in the range of [0%, 10%]; The grid area is divided into medium load area, the percentage of fixed load is in the range of (30%, 50%], and the percentage of electric vehicle charging load is in the range of (10%, 20%); The grid area is divided into uniform load areas, the fixed load percentage range is (50%, 70%], and the electric vehicle charging load percentage range is (20%, 30%]; The grid area divided into heavy-load areas has a fixed load percentage range of (70%, 100%] and an electric vehicle charging load percentage range of (30%, 40%]; For the grid areas divided into overload areas, the percentage of fixed load is in the range of >100%, and the percentage of electric vehicle charging load is in the range of >40%. S402, according to the load classification level standard, the grid of the grid division result is filled with a grid color, the color of the overload area is filled with: R: 90%, G: 10%, B: 10%; the color of the heavy load area is filled with: R: 100%, G: 0%, B: 0%; the color of the even load area is filled with: R: 80%, G: 20%, B: 20%; the color of the medium load area is filled with: R: 70%, G: 30%, B: 30%; the color of the light load area is filled with: R: 100%, G: 30%, B: 0%; S403, according to the above steps, obtain the grid area load division result.

7. The electric vehicle charging station planning method based on load and vehicle flow distribution according to claim 1 is characterized in that: Step S5 is specifically as follows: S501. Based on the historical traffic flow data, the grid areas in the distribution network grid optimization map are divided into traffic thermal levels; the traffic thermal levels include completely unblocked areas, basically unblocked areas, normal operating areas, slow-moving areas, congested areas, and severely congested areas. The traffic thermal level division is performed as follows: The grid area is divided into completely unblocked areas, and the traffic thermal value range is [0,50] vehicles / hour; The grid area is divided into basically unblocked areas, and the traffic thermal value range is (50,200] vehicles / hour; The grid area is divided into the normal operation area, and the traffic thermal value range is (200,500] vehicles / hour; The grid area divided into slow-moving areas has a traffic heat value range of (500,1000] vehicles / hour; The grid area divided into congested areas has a traffic thermal value range of (1000,2000] vehicles / hour; For grid areas classified as heavily congested areas, the traffic thermal value range is >2000 vehicles / hour; S502. According to the traffic thermal classification level standard, the grid division result is filled with grid colors. The chromaticity of the heavily congested area is: R: 43%, G: 0%, B: 100%; the chromaticity of the congested area is: R: 29%, G: 0%, B: 51%; the chromaticity of the slow-moving area is: R: 0%, G: 25%, B: 50%; the chromaticity of the normal driving area is: R: 0%, G: 50%, B: 100%; the chromaticity of the basically unblocked area is: R: 0%, G: 100%, B: 70%; the chromaticity of the completely unblocked area is: R: 0%, G: 40%, B: 60%; S503. According to the above steps, the grid area traffic thermal division result is obtained.

8. The electric vehicle charging station planning method based on load and vehicle flow distribution according to claim 1 is characterized in that: Step S6 is specifically as follows: S601, according to the grid area load division result and the grid area traffic thermal division result, based on the division result of each grid area in the area to be planned, superimpose the grid area load division result corresponding to each grid area and the chromaticity of the grid area traffic thermal division result; S602. Based on the color superposition results of each grid area, plan the location and capacity of the charging piles according to the following standards:

Images