Method and system for medium- and long-term prediction of electric vehicle charging load in consideration of orderly charging strategy

Abstract

Disclosed in the present invention are a method and system for medium- and long-term prediction of an electric vehicle charging load in consideration of an orderly charging strategy. The method comprises: predicting the number of electric vehicles that are present within a region, collecting the average daily mileage and power consumption of the electric vehicles, and calculating the charging electric quantity of the electric vehicles in a predicted year; on the basis of a Monte Carlo simulation method and travel rules of the electric vehicles, determining a typical charging load curve of electric vehicles in a region to be predicted; in consideration of the impact of an orderly charging strategy, on the basis of an elasticity coefficient and a charging price in the predicted year, correcting the typical charging load curve of the electric vehicles in the region to be predicted; and on the basis of the charging electric quantity of the electric vehicles in the predicted year and a corrected typical charging load curve of the electric vehicles in a year to be predicted, obtaining a final electric vehicle charging load of the region to be predicted in the year to be predicted. The dynamic response capability for prediction is improved, and the impact of peak loads on a power grid is reduced. The problem of a power grid being overloaded during peak periods is prevented, and the operation stability and efficiency of the power grid are improved. The utilization rate of charging infrastructure is improved.

Description

A medium- and long-term prediction method and system for electric vehicle charging load considering orderly charging strategy Technical Field

[0001] The present invention relates to the technical field of electric vehicle charging, and in particular to a method and system for predicting the mid- to long-term charging load of an electric vehicle taking into account an orderly charging strategy. Background Art

[0002] With the continuous growth of global energy demand and increasing environmental awareness, electric vehicle (EV) technology has experienced rapid development. Due to their environmentally friendly and energy-saving characteristics, EVs have gradually become a key solution to the energy crisis and reduce greenhouse gas emissions. In recent years, with the rapid expansion of the EV market, the construction of charging infrastructure has also rapidly followed. However, the unpredictability and volatility of EV charging loads have posed new challenges to the stable operation of the power grid. Traditional charging load forecasting methods, which rely primarily on historical data and simple linear models, are unable to accurately reflect the charging behavior of EV users and the changing trends in future charging demand. These methods often fail to account for complex factors such as charging prices and the elastic response of user charging behavior, resulting in significant deviations between the forecast results and actual load.

[0003] Existing technologies have many shortcomings in predicting electric vehicle charging loads. First, most methods fail to fully consider the dynamic changes in electric vehicle charging behavior and ignore the temporal and spatial distribution characteristics of the charging load. Second, existing prediction methods are mostly based on static models and lack analysis of charging price fluctuations and user behavior responses, which may lead to grid overload problems during peak hours. In addition, in the charging load prediction process, existing technologies often lack effective data processing and analysis methods and cannot accurately capture the randomness and complexity of charging behavior. In contrast, the orderly charging strategy proposed in the present invention, by combining the Monte Carlo simulation method and the laws of user charging behavior, can more accurately predict the charging load of electric vehicles and dynamically adjust the charging load according to the charging price and elasticity coefficient in the forecast year. This not only effectively reduces the pressure on the power grid during peak hours, but also improves the utilization efficiency of charging infrastructure.

[0004] The existing technology has the above-mentioned problems in the prediction of electric vehicle charging load. The present invention proposes a medium- and long-term prediction method for electric vehicle charging load considering an orderly charging strategy. By predicting the number of electric vehicles in a region, collecting the average daily mileage and power consumption of electric vehicles, and calculating the charging power of electric vehicles in the prediction year; based on the Monte Carlo simulation method and the travel pattern of electric vehicles, the typical charging load curve of electric vehicles in the predicted area is determined; considering the impact of the orderly charging strategy, the typical charging load curve of electric vehicles in the predicted area is corrected according to the elasticity coefficient and the charging price in the predicted year; based on the charging power of electric vehicles in the prediction year and the corrected charging load typical curve of electric vehicles in the prediction year, the final charging load of electric vehicles in the predicted area in the predicted year is obtained.

[0005] Summary of the Invention

[0006] In view of the above-mentioned problems, the present invention is proposed.

[0007] Therefore, the technical problem solved by the present invention is that the existing charging load prediction methods are mostly based on static models, which have insufficient prediction accuracy, lack of dynamic response, limited data processing capabilities, and the problem of power grid overload during peak hours.

[0008] To solve the above technical problems, the present invention provides the following technical solution: a method for medium- and long-term prediction of electric vehicle charging load considering an orderly charging strategy, comprising:

[0009] Predict the number of electric vehicles in the region, collect the average daily mileage and power consumption of electric vehicles, and calculate the charging power of electric vehicles in the forecast year;

[0010] Based on the Monte Carlo simulation method and the travel patterns of electric vehicles, the typical charging load curve of electric vehicles in the predicted area is determined;

[0011] Taking into account the impact of the orderly charging strategy, the typical charging load curve of electric vehicles in the forecast area is modified according to the elasticity coefficient and the charging price in the forecast year;

[0012] According to the charging power of electric vehicles in the forecast year and the typical curve of the corrected charging load of electric vehicles in the forecast year, the final charging load of electric vehicles in the forecast area in the forecast year is obtained.

[0013] As a preferred solution of the medium- and long-term prediction method of electric vehicle charging load considering an orderly charging strategy described in the present invention, wherein: the number of electric vehicles in a region is predicted, the average daily mileage and power consumption of electric vehicles are collected, and the charging power of electric vehicles in the predicted year is calculated, where electric vehicles include passenger cars, buses, and special-purpose vehicles;

[0014] Based on the economic and social development planning, urban and rural planning, comprehensive land utilization planning, environmental protection planning, smart transportation planning, cultural and ecotourism planning, electric vehicle and charging service industry planning, and smart grid planning of the region to be predicted, the number of electric vehicles in the region to be predicted in the forecast year is predicted;

[0015] Obtain the average daily mileage of electric vehicles and the power consumption of electric vehicles;

[0016] The charging capacity of electric vehicle charging load in the forecast year is calculated as follows: TC = 365 × (N1 × A1 × B1 + N2 × A2 × B2 + N3 × A3 × B3)

[0017] The formula for charging power of passenger cars in the forecast year is: TC1=365×N1×A1×B1

[0018] The formula for the charging capacity of buses in the forecast year is: TC2=365×N2×A2×B2

[0019] The formula for charging power of special-purpose vehicles in the forecast year is: TC3=365×N3×A3×B3

[0020] Among them, N1 represents the number of passenger cars in the forecast area in the forecast year, A1 represents the average daily mileage of passenger cars, B1 represents the power consumption per unit mileage of passenger cars, and TC1 represents the total charging power of passenger cars in the forecast year. N2 represents the number of buses in the forecast area in the forecast year, A2 represents the average daily mileage of buses, B2 represents the power consumption per unit mileage of buses, and TC2 represents the total charging power of buses in the forecast year. N3 represents the number of special-purpose vehicles in the forecast area in the forecast year, B3 represents the power consumption per unit mileage of special-purpose vehicles, and TC3 represents the total charging power of special-purpose vehicles in the forecast year.

[0021] As a preferred solution of the medium- and long-term prediction method of electric vehicle charging load considering an orderly charging strategy described in the present invention, based on the Monte Carlo simulation method and the travel pattern of electric vehicles, the typical curve of the charging load of electric vehicles in the area to be predicted is determined, including:

[0022] The daily driving distance s1 of passenger vehicles in the predicted area conforms to the log-normal distribution, and the probability density function is expressed as:

[0023] Among them, f s1 represents the probability density function of the daily driving distance s1 of passenger cars in the predicted area, s1 represents the specific value of the daily driving distance of passenger cars, σ s1 represents the logarithmic standard deviation of the daily driving distance s1 of passenger cars, μ s1 represents the logarithmic mean of the daily distance traveled by passenger cars s1;

[0024] The charging start time t1 of passenger vehicles in the predicted area adopts normal distribution, which is expressed as follows:

[0025] Among them, f t1 represents the probability distribution of starting charging at a specific time t1, where t1 represents the specific time when the passenger car starts charging, σ t1 represents the standard deviation of the time t1 at which the passenger car starts charging, μ t1 represents the mean value of the charging start time t1 of the passenger car;

[0026] The travel pattern parameter μ of passenger cars in the area to be predicted s1 , σ s1 、μ t1 and σ t1 Based on the historical charging data and the Monte Carlo simulation method, under the premise that the number of passenger cars in the predicted area in the predicted year is N1, the typical charging load curve of passenger cars in the predicted area is obtained. C1=[C1(1), C1(2), C1(3),……, C1(n)], where C1(1), C1(2), C1(3), C1(n) represent the charging load power of passenger cars in the predicted area at the first time point, the second time point, the third time point and the nth time point of the day respectively.

[0027] The daily travel routes of buses in the predicted area are relatively fixed, and the probability density function of the daily travel distance s2 is expressed as:

[0028] Among them, f s2 represents the daily travel distance of buses in the area to be predicted, s2 represents the specific value of the daily travel distance of buses, σ s2 : Standard deviation of daily bus travel distance s2, μ s2 represents the mean value of the daily travel distance s2 of buses;

[0029] The starting charging time t2 of buses in the predicted area should be evenly distributed within a given time period, and the formula is expressed as:

[0030] Among them, t2 represents the specific time when the bus starts charging, a represents the lower limit of the time when the bus starts charging, and f s2 represents the probability density function of the time t2 at which buses start charging in the test area, where t2 represents the specific time at which buses start charging;

[0031] The travel pattern parameter μ of passenger cars in the area to be predicted s2 , σ s2 , a and b are obtained through historical charging data;

[0032] Using the Monte Carlo simulation method, the number of buses in the predicted area in the predicted year, C2(1), C2(2), C2(3), C2(n) respectively represent the charging load power of buses in the predicted area at the first time point, the second time point, the third time point and the nth time point of the day;

[0033] The daily mileage of special vehicles in the forecast area fluctuates greatly, and the probability density function of the daily mileage s3 is expressed as: s3 =λe -λs3

[0034] Among them, f s3 represents the probability density function of the daily driving distance s3 of the special-purpose vehicle in the area to be predicted, s3 represents the specific value of the daily driving distance of the special-purpose vehicle, and λ represents the distribution parameter;

[0035] The charging start time t3 of the special-purpose vehicle in the predicted area is expressed by the normal distribution formula:

[0036] Among them, f t3 represents the probability density function of the time t3 at which the special-purpose vehicle starts charging in the area to be predicted, where t3 represents the specific time at which the special-purpose vehicle starts charging, and σ t3 represents the standard deviation of the time t3 when the special vehicle starts charging, μ t3 represents the mean value of the charging start time t3 of the special purpose vehicle;

[0037] The travel pattern parameters λ and μ of special vehicles in the area to be predicted t3 and σ t3 Obtained through historical charging data;

[0038] Using the Monte Carlo simulation method, under the premise that the number of special-purpose vehicles in the forecast area in the forecast year is N3, the typical charging load curve of special-purpose vehicles in the forecast area is obtained;

[0039] C3(1), C3(2), C3(3), and C3(n) represent the charging load power of special vehicles in the predicted area at the first time point, the second time point, the third time point, and the nth time point throughout the day, respectively.

[0040] As a preferred embodiment of the medium- and long-term prediction method for electric vehicle charging load considering an orderly charging strategy described in the present invention, the method includes: considering the impact of the orderly charging strategy, and modifying the typical curve of electric vehicle charging load in the predicted area according to the elasticity coefficient and the charging price in the predicted year, including establishing a Nash equilibrium model for passenger cars in the predicted area participating in orderly charging, and determining the elasticity coefficient α1 of passenger cars in the predicted area, which is expressed as follows:

[0041] Among them, F1 represents the main profit of passenger cars after participating in the orderly charging strategy, P1 represents the charging electricity price at the i-th time point of the day, Ch1(i) represents the charging power at the i-th time point of the day, Δt represents the interval time between adjacent time points, and FI1 represents the average daily charging cost of passenger cars without participating in orderly charging;

[0042] Under the orderly charging strategy of passenger cars in the forecasted area, the main profit formula that the power grid can obtain is expressed as: F2=T 1a max(C1(i)-Ch1(i))+α 1b max(C1(i)-Ch1(i)) β 1b +δ 1c max(C1(i)-Ch1(i))

[0043] Among them, F2 represents the main profit of the power grid after the passenger car participates in the orderly charging strategy, T 1a represents the cost per unit capacity of the transformer, max(C1(i)-Ch1(i)) represents the maximum charging power reduction at the i-th time point throughout the day, and α 1b represents the daily equivalent utilization coefficient of the distribution line where the charging station is located, β 1b represents the loss cost per unit power of the distribution line, δ 1c represents the benefits of delaying grid reconstruction projects under unit power reduction;

[0044] In the process of cooperation, all parties involved in the game hope to maximize their own interests. The Nash equilibrium model formula is expressed as:

[0045] Among them, F represents the total profit of the system, max(F1) represents the maximum profit obtained by the passenger car through the orderly charging strategy, max(F2) represents the maximum profit obtained by the power grid through the orderly charging strategy of the passenger car, F1 represents the profit after the passenger car participates in the orderly charging strategy, and F2 represents the profit of the power grid after the passenger car participates in the orderly charging strategy. F1>0 means that the profit of the passenger car must be positive, and F2>0 means that the profit of the power grid must be positive.

[0046] As a preferred embodiment of the medium- and long-term prediction method for electric vehicle charging load considering an orderly charging strategy described in the present invention, wherein: considering the influence of the orderly charging strategy, according to the elasticity coefficient and the charging price in the forecast year, correcting the typical charging load curve of electric vehicles in the forecast area further includes solving the established Nash equilibrium model to obtain the orderly charging power of passenger vehicles in the forecast area. In the Nash equilibrium model, F1 and F2 are simultaneously affected by the orderly charging power sequence of passenger vehicles at different times, which is a non-convex nonlinear optimization problem. It is converted into two sub-problems for solution. One sub-problem is to maximize the total benefit F3 brought by the benefits of orderly charging of passenger vehicles and the benefits obtained by the power grid. The formula is expressed as follows:

[0047] Among them, F3 represents the comprehensive profit of passenger cars and power grid after orderly charging strategy, P1(i) represents the charging electricity price at the i-th time point of the day, Ch1(i) represents the charging power at the i-th time point of the day, Δt represents the interval between adjacent time points, FI1 represents the average daily charging cost of passenger cars without participating in orderly charging, T 1a represents the cost per unit capacity of the transformer, max(C1(i)-Ch1(i)) represents the maximum charging power reduction at the i-th time point throughout the day, and α 1b represents the daily equivalent utilization coefficient of the distribution line where the charging station is located, β 1b represents the loss cost per unit power of the distribution line, δ 1c represents the benefits of delaying grid reconstruction projects under unit power reduction;

[0048] Another sub-problem is the most reasonable distribution of the benefits brought by orderly charging of passenger vehicles and the benefits obtained by the power grid. The formula is expressed as:

[0049] Where F4 represents the logarithmic comprehensive profit of the system, F1(i) represents the charging electricity price at the i-th time point, Cl1(i) represents the charging power at the i-th time point, Δt represents the interval between adjacent time points, FI1 represents the average daily charging cost of passenger cars without participating in orderly charging, T 1a represents the cost per unit capacity of the transformer, c 1b represents the daily equivalent utilization coefficient of the distribution line, β 1b represents the loss cost per unit power of the distribution line, δ 1c It represents the benefits brought by delaying the grid reconstruction project under unit power reduction.

[0050] As a preferred embodiment of the medium- and long-term prediction method for electric vehicle charging load considering an orderly charging strategy described in the present invention, the method further includes: considering the influence of the orderly charging strategy, modifying the typical charging load curve of electric vehicles in the predicted area according to the elasticity coefficient and the charging price in the predicted year, and solving two sub-problems by a traditional convex optimization algorithm to ensure that both F3 and F4 reach their maximum values, completing the solution of the established Nash equilibrium model, and forming the charging power Ch1(i) of passenger cars in the predicted area at the i-th time point throughout the day after participating in orderly charging. The elasticity coefficient α1 of passenger cars in the predicted area is obtained by the formula expressed as follows:

[0051] Using Nash equilibrium theory, the elastic coefficient α3 of special vehicles in the forecast area is determined. The special vehicles in the forecast area can participate in the orderly charging strategy through the main profit F 31 To obtain benefits, the formula is:

[0052] Among them, F 31 represents the main profit of the special-purpose vehicle after participating in the orderly charging strategy, P3(i) represents the charging electricity price at the i-th time point of the day, Ch3(i) represents the charging power at the i-th time point of the day, Δt represents the interval time between adjacent time points, and FI3 represents the average daily charging cost of the special-purpose vehicle without participating in the orderly charging strategy;

[0053] Under the strategy of orderly charging of special vehicles in the predicted area, the power grid can obtain the main profit F 32 The main formula is: F 32 =T 3a max(C3(i)-Ch3(i))+α 3b max(C3(i)-Ch3(i)) β 3b +δ 3c max(C3(i)-Ch3(i))

[0054] Among them, F 32 represents the main profit of the grid after the special vehicles participate in the orderly charging strategy, T 3a represents the cost per unit capacity of the transformer, C3(i) represents the charging load curve value at the i-th time point throughout the day, Ch3(i) represents the charging power at the i-th time point throughout the day, α 3b represents the daily equivalent utilization coefficient of the distribution line where the charging station is located, β 3b represents the loss cost per unit power of the distribution line, δ 3c represents the benefits of delaying grid reconstruction projects under unit power reduction;

[0055] In the process of cooperation, all parties involved in the game hope to maximize their own interests. The Nash equilibrium model is:

[0056] Solve the optimization formula to form the charging power Ch3(i) of the special vehicles in the predicted area at the i-th time point of the day after participating in the orderly charging, and obtain the elastic coefficient α3 of the passenger cars in the predicted area:

[0057] According to the elastic coefficient α1 of passenger cars in the predicted area and the charging price in the predicted year, P1(1), P1(2), P1(3), and P1(n) represent the charging electricity price of passenger cars in the predicted area at the first time point, the second time point, the third time point, and the nth time point of the day respectively;

[0058] The typical curve of charging load of passenger vehicles in the predicted area is expressed as follows:

[0059] C lz =

[0060] [C 1z (1),C 1z (2),C 1z (3),......,C 1z (n)],C 1z (1),C 1z (2),C 1z (3),C 1z (n) represents the corrected charging load power of passenger vehicles in the predicted area at the first, second, third, and nth time points throughout the day, respectively;

[0061] Among them, C 1z (i) represents the adjusted charging load power of the passenger car at the i-th time point of the day, α1 represents the elasticity coefficient of the passenger car, max(P1) represents the maximum charging electricity price of the day, min(P1) represents the minimum charging electricity price of the day, P1(i) represents the charging electricity price at the i-th time point of the day, C1(i) ​​represents the original charging load power of the passenger car at the i-th time point of the day, and 1≤i≤n represents the range of time point i;

[0062] According to the elastic coefficient α3 of the special vehicles in the forecast area, the charging price P3 in the forecast year = [P3(1), P3(2), P3(3), ...., P3(n)], P3(1), P3(2), P3(3), P3(n) respectively represent the charging electricity price of the special vehicles in the forecast area at the first time point, the second time point, the third time point and the nth time point of the day;

[0063] The typical curve of charging load of special vehicles in the area to be predicted is expressed as: C3=[C3(1),C3(2),C3(3),......,C3(n)] C 3z =[C 3z (1),C 3z (2),C 3z (3),…….,C 3z (n)]

[0064] C 3z (1),C 3z (2),C 3z (3),C 3z (n) represents the corrected charging load power of the special vehicle in the predicted area at the 1st time point, the 2nd time point, the 3rd time point and the nth time point throughout the day respectively.

[0065] Among them, C 3z (i) represents the adjusted charging load power of the special vehicle at the i-th time point of the day, α3 represents the elasticity coefficient of the special vehicle, max(P3) represents the maximum charging electricity price of the whole day, min(P3) represents the minimum charging electricity price of the whole day, P3(i) represents the charging electricity price at the i-th time point of the day, C3(i) represents the original charging load power of the special vehicle at the i-th time point of the day, and 1≤i≤n represents the range of time point i.

[0066] As a preferred embodiment of the medium- and long-term prediction method for electric vehicle charging load considering an orderly charging strategy described in the present invention, the method comprises: obtaining the final electric vehicle charging load for the forecasted area in the forecasted year based on the charging power of electric vehicles in the forecasted year and the typical curves of the corrected charging loads of various types of electric vehicles in the forecasted year; and calculating the passenger vehicle charging load prediction coefficient F1 for the forecasted area in the forecasted year, specifically:

[0067] Calculate the passenger car charging load prediction coefficient F2 for the predicted area in the predicted year, specifically:

[0068] Calculate the special vehicle charging load prediction coefficient F3 for the forecast area in the forecast year, specifically:

[0069] The final electric vehicle charging load for the forecasted area in the forecasted year is LP=[LP(1),LP(2),LP(3),.......,LP(n)], specifically: LP(i)=F1×C 1z (i)+F2×C2(i)+F3×C 3z (i)1≤i≤n

[0070] Among them, F1 represents the charging load prediction coefficient of passenger cars, N1 represents the number of passenger cars in the prediction area in the prediction year, A1 represents the average daily mileage of passenger cars, F2 represents the charging load prediction coefficient of buses, N2 represents the number of buses in the prediction area in the prediction year, A2 represents the average daily mileage of buses, F3 represents the charging load prediction coefficient of special-purpose vehicles, N3 represents the number of special-purpose vehicles in the prediction area in the prediction year, and A3 represents the average daily mileage of special-purpose vehicles.

[0071] A medium- and long-term forecasting system for electric vehicle charging load considering an orderly charging strategy is proposed, wherein:

[0072] The prediction module predicts the number of electric vehicles in the region, collects the average daily mileage and power consumption of electric vehicles, and calculates the charging power of electric vehicles in the predicted year;

[0073] The determination module determines the typical charging load curve of electric vehicles in the predicted area based on the Monte Carlo simulation method and the travel patterns of electric vehicles;

[0074] The correction module considers the impact of the orderly charging strategy and corrects the typical charging load curve of electric vehicles in the forecast area based on the elasticity coefficient and the charging price in the forecast year;

[0075] The result module obtains the final charging load of electric vehicles in the forecast area in the forecast year based on the charging power of electric vehicles in the forecast year and the typical curve of the corrected charging load of electric vehicles in the forecast year.

[0076] A computer device comprises: a memory and a processor; the memory stores a computer program, wherein the processor implements the steps of any one of the methods of the present invention when executing the computer program.

[0077] A computer-readable storage medium stores a computer program thereon, wherein the computer program, when executed by a processor, implements the steps of any one of the methods of the present invention.

[0078] The beneficial effects of the present invention are as follows: The present invention provides a medium- and long-term prediction method for electric vehicle charging load that takes into account an orderly charging strategy. This method combines Monte Carlo simulation with the travel patterns of electric vehicle users, enabling more accurate prediction of electric vehicle charging loads. It dynamically modifies the typical charging load curve, improving the dynamic response of the prediction and reducing the impact of peak loads on the power grid. This prevents grid overloads during peak hours and improves operational stability and efficiency. By optimizing charging time distribution and load management, it increases the utilization rate of charging infrastructure. This prevents excessive congestion at charging stations during peak hours and waste of resources during off-peak hours, achieving efficient resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0080] FIG1 is an overall flow chart of a method for medium- and long-term prediction of electric vehicle charging load taking into account an orderly charging strategy, provided by a first embodiment of the present invention. DETAILED DESCRIPTION

[0081] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without creative work should fall within the scope of protection of the present invention.

[0082] Example 1

[0083] 1 , which shows an embodiment of the present invention, provides a method for medium- and long-term prediction of electric vehicle charging load considering an orderly charging strategy, including:

[0084] S1: Predict the number of electric vehicles in the region, collect the average daily mileage and power consumption of electric vehicles, and calculate the charging power of electric vehicles in the predicted year.

[0085] Electric vehicles include passenger cars, buses and special-purpose vehicles.

[0086] Based on the economic and social development planning, urban and rural planning, comprehensive land utilization planning, environmental protection planning, smart transportation planning, cultural and ecological tourism planning, electric vehicle and charging service industry planning and smart grid planning of the region to be predicted, the number of electric vehicles in the region to be predicted in the forecast year is predicted.

[0087] Get the average daily mileage of electric vehicles and the power consumption of electric vehicles.

[0088] The charging capacity of electric vehicle charging load in the forecast year is calculated as follows: TC = 365 × (N1 × A1 × B1 + N2 × A2 × B2 + N3 × A3 × B3)

[0089] The formula for charging power of passenger cars in the forecast year is: TC1=365×N1×A1×B1

[0090] The formula for the charging capacity of buses in the forecast year is: TC2=365×N2×A2×B2

[0091] The formula for charging power of special-purpose vehicles in the forecast year is: TC3=365×N3×A3×B3

[0092] Among them, N1 represents the number of passenger cars in the forecast area in the forecast year, A1 represents the average daily mileage of passenger cars, B1 represents the power consumption per unit mileage of passenger cars, and TC1 represents the total charged power of passenger cars in the forecast year. N2 represents the number of buses in the forecast area in the forecast year, A2 represents the average daily mileage of buses, B2 represents the power consumption per unit mileage of buses, and TC2 represents the total charged power of buses in the forecast year. N3 represents the number of special-purpose vehicles in the forecast area in the forecast year, B3 represents the power consumption per unit mileage of special-purpose vehicles, and TC3 represents the total charged power of special-purpose vehicles in the forecast year.

[0093] By collecting and analyzing detailed planning data, vehicle ownership, mileage, and power consumption, we calculate the amount of electricity required to charge electric vehicles in the forecast year. This approach takes into account multiple factors and variables, ensuring the accuracy and reliability of the forecast results.

[0094] Through detailed data collection and analysis, future EV charging demand can be more accurately predicted, providing reliable data support for grid planning and management. This method considers multiple factors, such as vehicle type, mileage, and power consumption, resulting in more comprehensive and detailed predictions. These predictions can be used to guide charging infrastructure development and grid load management, avoiding resource waste and grid overload. Accurate load forecasting can optimize charging times and load distribution, improving the utilization efficiency of the grid and charging facilities. The data and predictions can inform government and relevant agencies' policy formulation, promoting the healthy development of the electric vehicle industry.

[0095] S2: Based on the Monte Carlo simulation method and the travel patterns of electric vehicles, determine the typical charging load curve of electric vehicles in the predicted area.

[0096] The daily driving distance s1 of passenger vehicles in the predicted area conforms to the log-normal distribution, and the probability density function is expressed as:

[0097] Among them, f s1 represents the probability density function of the daily driving distance s1 of passenger cars in the predicted area, s1 represents the specific value of the daily driving distance of passenger cars, σ s1 represents the logarithmic standard deviation of the daily driving distance s1 of passenger cars, μ s1 represents the logarithmic mean of the daily distance traveled by passenger cars, s1.

[0098] The charging start time t1 of passenger vehicles in the predicted area adopts normal distribution, which is expressed as follows:

[0099] Among them, f t1 represents the probability distribution of starting charging at a specific time t1, where t1 represents the specific time when the passenger car starts charging, σ t1 represents the standard deviation of the time t1 at which the passenger car starts charging, μ t1 It represents the average value of the charging start time t1 of passenger cars.

[0100] The travel pattern parameter μ of passenger cars in the area to be predicted s1 , σ s1 、μ t1 and σ t1 Based on the historical charging data and the Monte Carlo simulation method, under the premise that the number of passenger cars in the predicted area in the predicted year is N1, the typical charging load curve of passenger cars in the predicted area is obtained. C1 = [C1(1), C1(2), C1(3), ..., C1(n)], where C1(1), C1(2), C1(3), C1(n) represent the charging load power of passenger cars in the predicted area at the first time point, the second time point, the third time point and the nth time point of the day respectively.

[0101] The daily travel routes of buses in the predicted area are relatively fixed, and the probability density function of the daily travel distance s2 is expressed as:

[0102] Among them, f s2 represents the daily travel distance of buses in the area to be predicted, s2 represents the specific value of the daily travel distance of buses, σ s2 : Standard deviation of daily bus travel distance s2, μ s2 It represents the mean daily travel distance s2 of buses.

[0103] The starting charging time t2 of buses in the predicted area should be evenly distributed within a given time period, and the formula is expressed as:

[0104] Among them, t2 represents the specific time when the bus starts charging, a represents the lower limit of the time when the bus starts charging, and f s2 It represents the probability density function of the time t2 at which the buses start charging in the test area, where t2 represents the specific time at which the buses start charging.

[0105] The travel pattern parameter μ of passenger cars in the area to be predicted s2 , σ s2 , a and b are obtained through historical charging data.

[0106] Using the Monte Carlo simulation method, the number of buses in the predicted area in the predicted year, C2(1), C2(2), C2(3), C2(n) respectively represent the charging load power of buses in the predicted area at the 1st time point, the 2nd time point, the 3rd time point and the nth time point throughout the day.

[0107] The daily mileage of special vehicles in the forecast area fluctuates greatly, and the probability density function of the daily mileage s3 is expressed as: s3 =λe -λs3

[0108] Among them, f s3 It represents the probability density function of the daily driving distance s3 of the special-purpose vehicle in the area to be predicted, s3 represents the specific value of the daily driving distance of the special-purpose vehicle, and λ represents the distribution parameter.

[0109] The charging start time t3 of the special-purpose vehicle in the predicted area is expressed by the normal distribution formula:

[0110] Among them, f t3 represents the probability density function of the time t3 at which the special-purpose vehicle starts charging in the predicted area, t3 represents the specific time at which the special-purpose vehicle starts charging, σ t3 represents the standard deviation of the time t3 when the special vehicle starts charging, μ t3 It represents the average value of the charging start time t3 of the special-purpose vehicle.

[0111] The travel pattern parameters λ and μ of special vehicles in the area to be predicted t3 and σ t3 Obtained through historical charging data.

[0112] Using the Monte Carlo simulation method, under the premise that the number of special-purpose vehicles in the predicted area in the predicted year is N3, the typical charging load curve of special-purpose vehicles in the predicted area is obtained; C3(1), C3(2), C3(3), and C3(n) represent the charging load power of special-purpose vehicles in the predicted area at the 1st time point, the 2nd time point, the 3rd time point, and the nth time point throughout the day, respectively.

[0113] By using Monte Carlo simulation methods and lognormal, normal, uniform, and exponential distributions to describe EV driving distance and charging time, we can more accurately simulate EV charging behavior. These distribution functions capture the randomness and complexity of EV charging behavior, making the predictions closer to reality.

[0114] By simulating the actual charging behavior of electric vehicles, the accuracy of charging load forecasts can be improved, providing reliable data support for power grid planning and management. The method takes into account the distribution characteristics of different vehicle types, driving distances, and charging times, making the forecast results more comprehensive and detailed. By accurately predicting the charging load of electric vehicles, the load management of the power grid can be optimized, the pressure of peak loads on the power grid can be reduced, and the stability and efficiency of power grid operation can be improved. The forecast results can be used to guide the layout and construction of charging infrastructure, avoiding resource waste and power grid overload. The flexibility of the Monte Carlo simulation method allows for multiple simulations based on different input parameters and assumptions, thereby evaluating the charging load distribution under different scenarios and providing a variety of decision support.

[0115] S3: Considering the impact of the orderly charging strategy, the typical charging load curve of electric vehicles in the predicted area is modified according to the elasticity coefficient and the charging price in the predicted year.

[0116] A Nash equilibrium model for orderly charging of passenger cars in the predicted area is established, and the elastic coefficient α1 of passenger cars in the predicted area is determined. The formula is expressed as:

[0117] Among them, F1 represents the main profit of passenger cars after participating in the orderly charging strategy, P1 represents the charging electricity price at the i-th time point of the day, Ch1(i) represents the charging power at the i-th time point of the day, Δt represents the interval time between adjacent time points, and FI1 represents the average daily charging cost of passenger cars without participating in orderly charging.

[0118] Under the orderly charging strategy of passenger cars in the forecasted area, the main profit formula that the power grid can obtain is expressed as: F2=T 1a max(C1(i)-Ch1(i))+α 1b max(C1(i)-Ch1(i)) β 1b +δ 1c max(C1(i)-Ch1(i))

[0119] Among them, F2 represents the main profit of the power grid after the passenger car participates in the orderly charging strategy, T 1a represents the cost per unit capacity of the transformer, max(C1(i)-Ch1(i)) represents the maximum charging power reduction at the i-th time point throughout the day, and α 1b represents the daily equivalent utilization coefficient of the distribution line where the charging station is located, β 1b represents the loss cost per unit power of the distribution line, δ 1c It represents the benefits brought by delaying the grid reconstruction project under unit power reduction.

[0120] In the process of cooperation, all parties involved in the game hope to maximize their own interests. The Nash equilibrium model formula is expressed as:

[0121] Among them, F represents the total profit of the system, max(F1) represents the maximum profit obtained by the passenger car through the orderly charging strategy, max(F2) represents the maximum profit obtained by the power grid through the orderly charging strategy of the passenger car, F1 represents the profit after the passenger car participates in the orderly charging strategy, and F2 represents the profit of the power grid after the passenger car participates in the orderly charging strategy. F1>0 means that the profit of the passenger car must be positive, and F2>0 means that the profit of the power grid must be positive.

[0122] Solving the established Nash equilibrium model to obtain the orderly charging power of passenger vehicles in the predicted area includes: In the Nash equilibrium model, F1 and F2 are simultaneously affected by the orderly charging power sequence of passenger vehicles at different times. It is a non-convex nonlinear optimization problem, which is converted into two sub-problems to be solved. One sub-problem is to maximize the total benefit F3 brought by the benefits of orderly charging of passenger vehicles and the benefits obtained by the power grid. The formula is expressed as:

[0123] Among them, F3 represents the comprehensive profit of passenger cars and power grid after orderly charging strategy, P1(i) represents the charging electricity price at the i-th time point of the day, Ch1(i) represents the charging power at the i-th time point of the day, Δt represents the interval between adjacent time points, FI1 represents the average daily charging cost of passenger cars without participating in orderly charging, T 1a represents the cost per unit capacity of the transformer, max(C1(i)-Ch1(i)) represents the maximum charging power reduction at the i-th time point throughout the day, and α 1b represents the daily equivalent utilization coefficient of the distribution line where the charging station is located, β 1b represents the loss cost per unit power of the distribution line, δ 1c It represents the benefits brought by delaying the grid reconstruction project under unit power reduction.

[0124] Another sub-problem is the most reasonable distribution of the benefits brought by orderly charging of passenger vehicles and the benefits obtained by the power grid. The formula is expressed as:

[0125] Where F4 represents the logarithmic comprehensive profit of the system, F1(i) represents the charging electricity price at the i-th time point, Cl1(i) represents the charging power at the i-th time point, Δt represents the interval between adjacent time points, FI1 represents the average daily charging cost of passenger cars without participating in orderly charging, T 1a represents the cost per unit capacity of the transformer, c 1b represents the daily equivalent utilization coefficient of the distribution line, β 1b represents the loss cost per unit power of the distribution line, δ1c It represents the benefits brought by delaying the grid reconstruction project under unit power reduction.

[0126] The two sub-problems can be solved by the traditional convex optimization algorithm to ensure that both F3 and F4 reach their maximum values, and the solution of the established Nash equilibrium model is completed. The charging power Ch1(i) of the passenger cars in the predicted area at the i-th time point of the day after participating in orderly charging is formed. The elastic coefficient α1 of the passenger cars in the predicted area is expressed as follows:

[0127] Using Nash equilibrium theory, the elastic coefficient α3 of special vehicles in the forecast area is determined. The special vehicles in the forecast area can participate in the orderly charging strategy through the main profit F 31 To obtain benefits, the formula is:

[0128] Among them, F 31 represents the main profit of special-purpose vehicles after participating in the orderly charging strategy, P3(i) represents the charging electricity price at the i-th time point of the day, Ch3(i) represents the charging power at the i-th time point of the day, Δt represents the interval time between adjacent time points, and FI3 represents the average daily charging cost of special-purpose vehicles without participating in orderly charging.

[0129] Under the strategy of orderly charging of special vehicles in the predicted area, the power grid can obtain the main profit F 32 The main formula is: F 32 =T 3a max(C3(i)-Ch3(i))+α 3b max(C3(i)-Ch3(i)) β 3b +δ 3c max(C3(i)-Ch3(i))

[0130] Among them, F 32 represents the main profit of the grid after the special vehicles participate in the orderly charging strategy, T 3a represents the cost per unit capacity of the transformer, C3(i) represents the charging load curve value at the i-th time point throughout the day, Ch3(i) represents the charging power at the i-th time point throughout the day, α 3b represents the daily equivalent utilization coefficient of the distribution line where the charging station is located, β 3b represents the loss cost per unit power of the distribution line, δ 3c It represents the benefits brought by delaying the grid reconstruction project under unit power reduction.

[0131] In the process of cooperation, all parties involved in the game hope to maximize their own interests. The Nash equilibrium model is:

[0132] Solve the optimization formula to form the charging power Ch3(i) of the special vehicles in the predicted area at the i-th time point of the day after participating in the orderly charging, and obtain the elastic coefficient α3 of the passenger cars in the predicted area:

[0133] According to the elasticity coefficient α1 of passenger cars in the predicted area and the charging price in the predicted year, P1(1), P1(2), P1(3), and P1(n) represent the charging electricity prices of passenger cars in the predicted area at the 1st time point, the 2nd time point, the 3rd time point, and the nth time point throughout the day, respectively.

[0134] The typical curve of charging load of passenger vehicles in the predicted area is expressed as follows:

[0135] C lz =[C 1z (1),C 1z (2),C 1z (3),......,C 1z (n)],C 1z (1),C 1z (2),C 1z (3),C 1z (n) represents the corrected charging load power of passenger cars in the predicted area at the 1st, 2nd, 3rd and nth time points throughout the day, respectively.

[0136] Among them, C 1z (i) represents the adjusted charging load power of the passenger car at the i-th time point of the day, α1 represents the elasticity coefficient of the passenger car, max(P1) represents the maximum charging electricity price of the day, min(P1) represents the minimum charging electricity price of the day, P1(i) represents the charging electricity price at the i-th time point of the day, C1(i) ​​represents the original charging load power of the passenger car at the i-th time point of the day, and 1≤i≤n represents the range of time point i.

[0137] According to the elasticity coefficient α3 of special vehicles in the predicted area and the charging price P3 in the predicted year = [P3(1), P3(2), P3(3),……., P3(n)], P3(1), P3(2), P3(3), P3(n) respectively represent the charging electricity price of special vehicles in the predicted area at the 1st time point, the 2nd time point, the 3rd time point and the nth time point throughout the day.

[0138] The typical curve of charging load of special vehicles in the area to be predicted is expressed as: C3=[C3(1),C3(2),C3(3),......,C3(n)] C 3z =[C 3z (1),C3z (2),C 3z (3),…….,C 3z (n)]

[0139] C 3z (1),C 3z (2),C 3z (3),C 3z (n) represents the corrected charging load power of the special vehicle in the predicted area at the 1st time point, the 2nd time point, the 3rd time point and the nth time point throughout the day respectively.

[0140] Among them, C 3z (i) represents the adjusted charging load power of the special vehicle at the i-th time point of the day, α3 represents the elasticity coefficient of the special vehicle, max(P3) represents the maximum charging electricity price of the whole day, min(P3) represents the minimum charging electricity price of the whole day, P3(i) represents the charging electricity price at the i-th time point of the day, C3(i) represents the original charging load power of the special vehicle at the i-th time point of the day, and 1≤i≤n represents the range of time point i.

[0141] Detailed mathematical models and formulas enable more accurate predictions of EV charging loads, improving forecast accuracy. Optimizing charging strategies can smooth load curves, reduce peak loads on the grid, and improve grid stability and efficiency. Dynamically adjusting charging strategies maximizes profits for both passenger vehicles and the grid, improving economic benefits. Accurately predicting and optimizing charging loads can avoid resource waste and rationally allocate charging infrastructure. Providing reliable data and analysis provides a scientific basis for policymaking and promotes the healthy development of the electric vehicle industry.

[0142] S4: Based on the charging power of electric vehicles in the forecast year and the typical curve of the corrected charging load of electric vehicles in the forecast year, the final charging load of electric vehicles in the forecast area in the forecast year is obtained.

[0143] Calculate the passenger car charging load prediction coefficient F1 for the forecast area in the forecast year, specifically:

[0144] Calculate the passenger car charging load prediction coefficient F2 for the predicted area in the predicted year, specifically:

[0145] Calculate the special vehicle charging load prediction coefficient F3 for the forecast area in the forecast year, specifically:

[0146] The final electric vehicle charging load for the forecast area in the forecast year is LP=[LP(1),LP(2),LP(3),.......,LP(n)], specifically: LP(i)=F1×C 1z (i)+F2×C2(i)+F3×C 3z (i) 1≤i≤n

[0147] Among them, F1 represents the charging load prediction coefficient of passenger cars, N1 represents the number of passenger cars in the prediction area in the prediction year, A1 represents the average daily mileage of passenger cars, F2 represents the charging load prediction coefficient of buses, N2 represents the number of buses in the prediction area in the prediction year, A2 represents the average daily mileage of buses, F3 represents the charging load prediction coefficient of special-purpose vehicles, N3 represents the number of special-purpose vehicles in the prediction area in the prediction year, and A3 represents the average daily mileage of special-purpose vehicles.

[0148] This forecast takes into account multiple factors, including vehicle ownership, mileage, and charging load, to provide more comprehensive and accurate predictions. By calculating charging load prediction coefficients for different vehicle types, it more accurately reflects the charging needs of each type of vehicle at different times, improving overall forecast accuracy. The forecast results can be used to optimize the planning and management of power grids and charging infrastructure, ensuring optimal resource allocation and avoiding grid overloads caused by peak loads. Accurate charging load forecasts can reduce charging infrastructure construction and operating costs, improve grid efficiency, and improve charging facility utilization. This provides reliable data support, providing a scientific basis for government and relevant agencies to formulate policies, and promote the healthy development of the electric vehicle industry.

[0149] On the other hand, this embodiment also provides a medium- and long-term prediction system for electric vehicle charging load considering an orderly charging strategy, which includes:

[0150] The prediction module predicts the number of electric vehicles in the region, collects the average daily mileage and power consumption of electric vehicles, and calculates the charging power of electric vehicles in the predicted year.

[0151] The determination module determines the typical charging load curve of electric vehicles in the predicted area based on the Monte Carlo simulation method and the travel patterns of electric vehicles.

[0152] The correction module considers the impact of the orderly charging strategy and corrects the typical charging load curve of electric vehicles in the predicted area according to the elasticity coefficient and the charging price in the predicted year.

[0153] The result module obtains the final charging load of electric vehicles in the forecast area in the forecast year based on the charging power of electric vehicles in the forecast year and the typical curve of the corrected charging load of electric vehicles in the forecast year.

[0154] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0155] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0156] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0157] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0158] Example 2

[0159] The experiment started with a certain city as the prediction area. According to the data from the city’s traffic management department, the number of electric vehicles in the predicted area is 30,000, including 20,000 passenger cars, 5,000 buses, and 5,000 special-purpose vehicles.

[0160] By collecting data from the city's electric vehicle monitoring system, we obtained the following average daily mileage and power consumption:

[0161] The average daily mileage of a passenger car is 50 kilometers, and the power consumption per unit mileage is 0.2kWh / km.

[0162] The average daily mileage of buses is 100 kilometers, and the power consumption per unit mileage is 0.8 kWh / km.

[0163] The average daily mileage of special-purpose vehicles is 80 kilometers, and the power consumption per unit mileage is 0.5kWh / km.

[0164] Annual charging power of a passenger car: 365×20,000×50×0.2=73,000,000kWh.

[0165] Annual charging capacity of buses: 365×5000×100×0.8=146,000,000kWh.

[0166] Annual charging capacity of special-purpose vehicles: 365×5000×80×0.5=73,000,000kWh.

[0167] Using Monte Carlo simulation methods, combined with electric vehicle travel patterns and charging behavior, we simulated the charging load curves for electric vehicles in different time periods within the city. Based on the elasticity coefficient of electric vehicle charging and the charging price in the forecast year, we modified the typical charging load curve. For example, if electric vehicle charging prices fluctuate during the forecast year, the price elasticity coefficient can be used to adjust the charging load for each time period to better reflect actual conditions. Based on this modified typical charging load curve and the charging power consumption during the forecast year, we calculated the final electric vehicle charging load for the forecasted area in the forecast year. The experimental data is shown in Table 1.

[0168] Table 1 Experimental data table

[0169] By adjusting charging loads, peak-hour charging loads were significantly reduced. For example, during the peak period from 4:00 PM to 8:00 PM, the charging load for passenger cars decreased from 18,000 kWh to 17,000 kWh, for buses from 20,000 kWh to 18,000 kWh, and for special-purpose vehicles from 15,000 kWh to 15,500 kWh. This adjustment effectively smoothed the load curve and reduced peak pressure on the power grid. After the adjustment, the charging load increased during off-peak hours. For example, between 12:00 AM and 4:00 AM, the charging load for passenger cars increased from 5,000 kWh to 5,500 kWh, and for special-purpose vehicles from 6,000 kWh to 6,500 kWh. This adjustment took advantage of off-peak periods and improved grid resource utilization. Based on the expected fluctuations in charging prices during the forecast year, charging loads were adjusted appropriately. For example, charging loads increased during periods with lower prices and decreased during periods with higher prices. This not only optimized charging costs for users but also improved the economic benefits of the power grid. By optimizing the charging load curve, peak loads are reduced, the need for grid expansion and transformation is postponed, and operating and maintenance costs are lowered. Monte Carlo simulation and the application of elasticity coefficients make the prediction results more closely aligned with actual charging behavior, improving the accuracy and reliability of the predictions.

[0170] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A medium- and long-term prediction method for electric vehicle charging load considering an orderly charging strategy, characterized in that: include: Predict the number of electric vehicles in the region, collect the average daily mileage and power consumption of electric vehicles, and calculate the charging power of electric vehicles in the forecast year; Based on the Monte Carlo simulation method and the travel patterns of electric vehicles, the typical charging load curve of electric vehicles in the predicted area is determined; Considering the impact of the orderly charging strategy, the typical curve of the charging load of electric vehicles in the forecast area is modified according to the elasticity coefficient and the charging price in the forecast year; According to the charging power of electric vehicles in the forecast year and the typical curve of the corrected charging load of electric vehicles in the forecast year, the final charging load of electric vehicles in the forecast area in the forecast year is obtained.

2. The method for predicting the charging load of an electric vehicle in the medium and long term considering the orderly charging strategy as claimed in claim 1, characterized in that: Predict the number of electric vehicles in the region, collect the average daily mileage and power consumption of electric vehicles, and calculate the charging power of electric vehicles in the forecast year. Electric vehicles include passenger cars, buses and special vehicles; According to the contents of the economic and social development planning, urban and rural planning, comprehensive land utilization planning, environmental protection planning, smart transportation planning, cultural and ecological tourism planning, electric vehicle and charging service industry planning and smart grid planning of the region to be predicted, the number of electric vehicles in the region to be predicted in the forecast year is predicted; Obtain the average daily mileage of electric vehicles and the power consumption of electric vehicles; The charging capacity of electric vehicle charging load in the forecast year is calculated and the formula is expressed as: TC=365×(N1×A1×B1+N2×A2×B2+N3×A3×B3) The formula for charging power of passenger cars in the forecast year is expressed as: TC1=365×N1×A1×B1 The formula for charging power of buses in the forecast year is expressed as: TC2=365×N2×A2×B2 The charging power formula of special-purpose vehicles in the forecast year is expressed as: TC3=365×N3×A3×B3 Among them, N1 represents the number of passenger cars in the forecast area in the forecast year, and A1 represents the number of passenger cars. The mean of the average daily mileage, B1 represents the power consumption per unit mileage of passenger cars, TC1 represents the total charging power of passenger cars in the forecast year, N2 represents the number of buses in the forecast area in the forecast year, A2 represents the average daily mileage of buses, B2 represents the power consumption per unit mileage of buses, TC2 represents the total charging power of buses in the forecast year, N3 represents the number of special-purpose vehicles in the forecast area in the forecast year, B3 represents the power consumption per unit mileage of special-purpose vehicles, and TC3 represents the total charging power of special-purpose vehicles in the forecast year.

3. A method for predicting the charging load of an electric vehicle in the medium and long term considering an orderly charging strategy as claimed in claim 2, characterized in that: Based on the Monte Carlo simulation method and the travel patterns of electric vehicles, the typical curve of charging load of electric vehicles in the predicted area is determined to include: The daily driving distance s1 of passenger cars in the predicted area conforms to the log-normal distribution, and the probability density function is expressed as: Among them, f s1 represents the probability density function of the daily driving distance s1 of passenger cars in the area to be predicted, s1 represents the specific value of the daily driving distance of passenger cars, σ s1 represents the logarithmic standard deviation of the daily driving distance s1 of passenger cars, μ s1 represents the logarithmic mean of the daily driving distance s1 of passenger cars; The charging start time t1 of passenger cars in the predicted area adopts normal distribution, and the formula is expressed as: Among them, f t1 represents the probability distribution of starting charging at a specific time t1, where t1 represents the specific time when the passenger car starts charging, σ t1 represents the standard deviation of the time t1 at which the passenger car starts charging, μ t1 It represents the mean value of the time t1 when the passenger car starts charging; The travel pattern parameter μ of passenger cars in the area to be predicted s1 , σ s1 , μ t1 and σ t1 Based on the historical charging data, using the Monte Carlo simulation method, under the premise that the number of passenger cars in the predicted area in the predicted year is N1, the typical charging load curve of passenger cars in the predicted area is obtained. C1=[C1(1), C1(2), C1(3), …, C1(n)], where C1(1), C1(2), C1(3), C1(n) represent the charging load power of passenger cars in the predicted area at the first time point, the second time point, the third time point and the nth time point of the day respectively. The daily travel routes of buses in the predicted area are relatively fixed, and the probability density function of the daily travel distance s2 is expressed as: Among them, f s2 represents the daily driving distance of buses in the area to be predicted, s2 represents the specific value of the daily driving distance of buses, σ s2 : Standard deviation of daily travel distance s2 of buses, μ s2 represents the mean value of the daily travel distance s2 of buses; The starting charging time t2 of buses in the predicted area should be evenly distributed within a given time period, and the formula is expressed as: Among them, t2 represents the specific time when the bus starts charging, a represents the lower limit of the time when the bus starts charging, and f s2 represents the probability density function of the time t2 at which the bus starts charging in the test area, where t2 represents the specific time at which the bus starts charging; The travel pattern parameter μ of the passenger cars in the area to be predicted s2 , σ s2 , a and b are obtained through historical charging data; Using the Monte Carlo simulation method, the number of buses in the predicted area in the predicted year, C2(1), C2(2), C2(3), C2(n) respectively represent the charging load power of buses in the predicted area at the first time point, the second time point, the third time point and the nth time point of the day; The daily mileage of special vehicles in the predicted area fluctuates greatly, and the probability density function of the daily mileage s3 is expressed as: f s3 =λe -λs3 Among them, f s3 represents the probability density function of the daily driving distance s3 of the special-purpose vehicle in the area to be predicted, s3 represents the specific value of the daily driving distance of the special-purpose vehicle, and λ represents the distribution parameter; The charging start time t3 of the special-purpose vehicle in the predicted area is expressed by the normal distribution formula: Among them, f t3 represents the probability density function of the time t3 at which the special-purpose vehicle starts charging in the area to be predicted, t3 represents the specific time at which the special-purpose vehicle starts charging, σ t3 represents the standard deviation of the charging start time t3 of the special vehicle, μ t3 It represents the mean value of the charging start time t3 of the special vehicle; The travel pattern parameters λ and μ of special vehicles in the area to be predicted t3 and σ t3 Obtained through historical charging data; Using the Monte Carlo simulation method, under the premise that the number of special-purpose vehicles in the forecast area in the forecast year is N3, the typical curve of charging load of special-purpose vehicles in the forecast area is obtained; C3(1), C3(2), C3(3), and C3(n) respectively represent the charging load power of special vehicles in the predicted area at the 1st time point, the 2nd time point, the 3rd time point, and the nth time point throughout the day.

4. A method for predicting the charging load of an electric vehicle in the medium and long term considering an orderly charging strategy as claimed in claim 3, characterized in that: Considering the impact of the orderly charging strategy, according to the elasticity coefficient and the charging price in the forecast year, the typical curve of the charging load of electric vehicles in the forecast area is modified, including establishing a Nash equilibrium model for passenger cars in the forecast area to participate in orderly charging, and determining the elasticity coefficient α1 of passenger cars in the forecast area. The formula is expressed as follows: Among them, F1 represents the main profit of passenger cars after participating in the orderly charging strategy, P1 represents the charging electricity price at the i-th time point of the day, Ch1(i) represents the charging power at the i-th time point of the day, Δt represents the interval time between adjacent time points, and FI1 represents the average daily charging cost of passenger cars without participating in orderly charging; Under the strategy of orderly charging of passenger cars in the predicted area, the main profit formula that the power grid can obtain is expressed as: F2=T 1a ·max(C1(i)-Ch1(i))+α 1b ·max(C1(i)-Ch1(i))·β 1b +δ 1c max(C1(i)-Ch1(i)) Among them, F2 represents the main profit of the power grid after the passenger car participates in the orderly charging strategy, T 1a represents the cost per unit capacity of the transformer, max(C1(i)-Ch1(i)) represents the maximum charging power reduction at the i-th time point throughout the day, and α 1b represents the daily equivalent utilization coefficient of the distribution line where the charging station is located, β 1b represents the loss cost per unit power of the distribution line, δ 1c It represents the benefits brought by delaying the grid reconstruction project under unit power reduction; In the process of cooperation, all parties involved in the game hope to maximize their own interests. The Nash equilibrium model formula is expressed as: Among them, F represents the total profit of the system, max(F1) represents the maximum profit obtained by the passenger car through the orderly charging strategy, max(F2) represents the maximum profit obtained by the power grid through the orderly charging strategy of the passenger car, F1 represents the profit of the passenger car after participating in the orderly charging strategy, F2 represents the profit of the power grid after the passenger car participates in the orderly charging strategy, F1>0 represents the profit of the passenger car must be positive, and F2>0 represents the profit of the power grid must be positive.

5. The method for medium- and long-term prediction of electric vehicle charging load considering an orderly charging strategy as claimed in claim 4, characterized in that: Considering the impact of the orderly charging strategy, according to the elasticity coefficient and the charging price in the forecast year, the typical curve of the charging load of electric vehicles in the forecast area is corrected, and the orderly charging power of passenger cars in the forecast area is obtained by solving the established Nash equilibrium model. In the Nash equilibrium model, F1 and F2 are simultaneously affected by the orderly charging power sequence of passenger cars at different times. It is a non-convex nonlinear optimization problem, which is converted into two sub-problems for solution. One sub-problem is that the total benefit F3 brought by the orderly charging of passenger cars and the benefits obtained by the power grid is the largest, and the formula is expressed as: Among them, F3 represents the comprehensive profit of passenger cars and power grids after orderly charging strategy, P1(i) represents the charging electricity price at the i-th time point of the day, Ch1(i) represents the charging power at the i-th time point of the day, Δt represents the interval time between adjacent time points, FI1 represents the average daily charging cost of passenger cars without orderly charging, T 1a represents the cost per unit capacity of the transformer, max(C1(i)-Ch1(i)) represents the maximum charging power reduction at the i-th time point throughout the day, and α 1b represents the daily equivalent utilization coefficient of the distribution line where the charging station is located, β 1b represents the loss cost per unit power of the distribution line, δ 1c It represents the benefits brought by delaying the grid reconstruction project under unit power reduction; Another sub-problem is the most reasonable distribution of the benefits brought by orderly charging of passenger cars and the benefits obtained by the power grid. The formula is expressed as: Where F4 represents the logarithmic comprehensive profit of the system, F1(i) represents the charging electricity price at the i-th time point, Cl1(i) represents the charging power at the i-th time point, Δt represents the interval time between adjacent time points, FI1 represents the average daily charging cost of a passenger car without orderly charging, T 1a represents the cost per unit capacity of the transformer, c 1b represents the daily equivalent utilization coefficient of the distribution line, β 1b represents the loss cost per unit power of the distribution line, δ 1c Represents the benefits brought by delaying the grid reconstruction project under unit power reduction.

6. A method for predicting the charging load of an electric vehicle in the medium and long term considering an orderly charging strategy as claimed in claim 5, characterized in that: Considering the impact of the orderly charging strategy, according to the elasticity coefficient and the charging price in the forecast year, the typical curve of the charging load of electric vehicles in the forecast area is modified. It also includes solving two sub-problems through the traditional convex optimization algorithm to ensure that both F3 and F4 reach the maximum value, completing the solution of the established Nash equilibrium model, and forming the charging power Ch1(i) of the passenger car in the forecast area at the i-th time point of the day after participating in the orderly charging. The elasticity coefficient α1 of the passenger car in the forecast area is expressed as follows: Using Nash equilibrium theory, the elasticity coefficient α3 of special vehicles in the predicted area is determined. The special vehicles in the predicted area can participate in the orderly charging strategy through the main profit F 31 To obtain revenue, the formula is: Among them, F 31 represents the main profit of the special vehicle after participating in the orderly charging strategy, P3(i) represents the charging electricity price at the i-th time point of the day, Ch3(i) represents the charging power at the i-th time point of the day, Δt represents the interval time between adjacent time points, and FI3 represents the average daily charging cost of the special vehicle without participating in orderly charging; Under the strategy of orderly charging of special vehicles in the predicted area, the power grid can obtain the main profit F 32 The main formula is expressed as: F 32 =T 3a max(C3(i)-Ch3(i))+α 3b ·max(C3(i)-Ch3(i))·β 3b +δ 3c max(C3(i)-Ch3(i)) Among them, F 32 represents the main profit of the power grid after the special vehicles participate in the orderly charging strategy, T 3a represents the cost per unit capacity of the transformer, C3(i) represents the charging load curve value at the i-th time point of the day, Ch3(i) represents the charging power at the i-th time point of the day, α 3b represents the daily equivalent utilization coefficient of the distribution line where the charging station is located, β 3b represents the loss cost per unit power of the distribution line, δ 3c It represents the benefits brought by delaying the grid reconstruction project under unit power reduction; In the process of cooperation, all parties involved in the game hope to maximize their own interests. The Nash equilibrium model is: Solve the optimization formula to form the charging power Ch3(i) of the special vehicles in the predicted area at the i-th time point of the day after participating in the orderly charging, and obtain the elasticity coefficient α3 of the passenger cars in the predicted area: According to the elasticity coefficient α1 of passenger cars in the predicted area and the charging price in the predicted year, P1(1), P1(2), P1(3), and P1(n) represent the charging electricity prices of passenger cars in the predicted area at the first time point, the second time point, the third time point, and the nth time point of the day, respectively; The typical curve of charging load of passenger cars in the predicted area is expressed as: C lz ＝ [C 1z (1),C 1z (2),C 1z (3),......,C 1z (n)],C 1z (1),C 1z (2),C 1z (3),C 1z (n) represents the corrected charging load power of passenger cars in the predicted area at the first time point, the second time point, the third time point and the nth time point of the day respectively; Among them, C 1z (i) represents the adjusted charging load power of the passenger car at the i-th time point of the day, α1 represents the elasticity coefficient of the passenger car, max(P1) represents the maximum charging price of the day, min(P1) represents the minimum charging price of the day, P1(i) represents the charging price at the i-th time point of the day, C1(i) ​​represents the original charging load power of the passenger car at the i-th time point of the day, and 1≤i≤n represents the range of time point i; According to the elasticity coefficient α3 of the special-purpose vehicles in the area to be predicted and the charging price P3 in the predicted year = [P3(1), P3(2), P3(3), ..., P3(n)], P3(1), P3(2), P3(3), P3(n) respectively represent the charging electricity price of the special-purpose vehicles in the area to be predicted at the first time point, the second time point, the third time point and the nth time point of the whole day; The typical curve of charging load of special vehicles in the area to be predicted is expressed as follows: C3=[C3(1),C3(2),C3(3),...,C3(n)] C 3z =[C 3z (1),C 3z (2),C 3z (3),…….,C 3z (n)] C 3z (1),C 3z (2),C 3z (3),C 3z (n) represents the corrected charging load power of the special vehicle in the predicted area at the 1st time point, the 2nd time point, the 3rd time point and the nth time point of the day respectively. Among them, C 3z (i) represents the adjusted charging load power of the special vehicle at the i-th time point of the day, α3 represents the elasticity coefficient of the special vehicle, max(P3) represents the maximum charging price of the whole day, min(P3) represents the minimum charging price of the whole day, P3(i) represents the charging price at the i-th time point of the day, C3(i) represents The original special vehicle charging load power at the i-th time point throughout the day, 1≤i≤n represents the range of time point i.

7. A method for predicting the charging load of an electric vehicle in the medium and long term considering an orderly charging strategy as claimed in claim 6, characterized in that: According to the charging power of electric vehicles in the forecast year and the typical curves of the corrected charging load of various types of electric vehicles in the forecast year, the final charging load of electric vehicles in the forecast area in the forecast year is obtained, and the charging load prediction coefficient F1 of passenger cars in the forecast area in the forecast year is obtained, which is specifically: Calculate the passenger car charging load prediction coefficient F2 for the predicted area in the predicted year, specifically: Calculate the prediction coefficient F3 of the special vehicle charging load in the predicted area and the predicted year, specifically: The final electric vehicle charging load LP of the predicted area in the predicted year is obtained as follows: LP(i)=F1×C 1z (i)+F2×C2(i)+F3×C 3z (i)1≤i≤n Among them, F1 represents the charging load prediction coefficient of passenger cars, N1 represents the number of passenger cars in the prediction area in the prediction year, A1 represents the average daily mileage of passenger cars, F2 represents the charging load prediction coefficient of buses, N2 represents the number of buses in the prediction area in the prediction year, A2 represents the average daily mileage of buses, F3 represents the charging load prediction coefficient of special-purpose vehicles, N3 represents the number of special-purpose vehicles in the prediction area in the prediction year, and A3 represents the average daily mileage of special-purpose vehicles.

8. A medium- and long-term prediction system for electric vehicle charging load taking into account an orderly charging strategy using the method as described in any one of claims 1 to 7, characterized in that: The prediction module predicts the number of electric vehicles in the region, collects the average daily mileage and power consumption of electric vehicles, and calculates the charging power of electric vehicles in the predicted year; A determination module determines the typical charging load curve of electric vehicles in the area to be predicted based on the Monte Carlo simulation method and the travel rules of electric vehicles; The correction module considers the impact of the orderly charging strategy and corrects the typical charging load curve of electric vehicles in the predicted area according to the elasticity coefficient and the charging price in the predicted year; The result module obtains the final charging load of electric vehicles in the forecast area in the forecast year based on the charging power of electric vehicles in the forecast year and the typical curve of the corrected charging load of electric vehicles in the forecast year.

9. A computer device comprising: A memory and a processor; the memory stores a computer program, characterized in that: when the processor executes the computer program, the processor implements the steps of a method for medium- and long-term prediction of electric vehicle charging load taking into account an orderly charging strategy.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a method for mid- to long-term prediction of electric vehicle charging load taking into account an orderly charging strategy are implemented.

Images