Electric vehicle charging load space-time distribution prediction method based on information physical fusion

Abstract

The invention belongs to the technical field of electric vehicle charging load prediction, and particularly relates to an electric vehicle charging load spatio-temporal distribution prediction method based on information physical fusion, which comprises the following steps of: generating parking duration, initial travel time, initial electric quantity, starting point position and terminal point position of a single electric vehicle; according to the method, the shortest travel path of the electric vehicle is obtained through shortest path planning, then the farthest drivable distance of the electric vehicle is calculated based on the remaining electric quantity when the electric vehicle generates the charging demand, then the final charging node is determined, and the shortest charging path is obtained through the shortest path planning. And then calculating the remaining capacity of the battery and the queuing waiting time when the electric vehicle arrives at the final charging node, determining the final charging node, the charging starting time and the charging power of the electric vehicle, and finally outputting a dynamic prediction result of the charging load spatial and temporal distribution of the electric vehicle. The method can improve the prediction precision of the dynamic prediction result of the time-space distribution of the charging load.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric vehicle charging load prediction, and in particular relates to a method for predicting the spatiotemporal distribution of electric vehicle charging load based on information-physics fusion. Background Art

[0002] As a green transportation tool with great development prospects, electric vehicles are an important means to tap the potential of carbon reduction in transportation, improve the level of electrification of transportation, and achieve carbon peak regulation and carbon neutrality goals. With the continuous growth of the number of electric vehicles, there will be huge demand for charging, which will stimulate the comprehensive deployment and rapid growth of charging infrastructure. However, compared with the continuously growing market demand, the comprehensive car-to-pile ratio still has a certain gap. Long charging time and poor charging experience are the main difficulties and challenges faced by existing electric vehicle charging technologies. Due to the lack of actual usage data of electric vehicles in the initial stage, the construction of charging infrastructure lacks support and guidance, resulting in unreasonable planning of charging station locations, mismatch between charging pile construction and actual demand ratio, resulting in very high charging waiting rates in some areas, and a large number of charging piles in remote areas are idle. This contradiction has become a considerable challenge facing the industry. Using computational intelligence to reasonably and efficiently plan the layout and capacity of charging stations to adapt to the actual spatiotemporal distribution of charging load demand is an urgent solution to solve the limited range of electric vehicles for electric vehicle users, improve the charging experience of electric vehicles, and increase the utilization rate of regional charging stations. This requires more accurate spatiotemporal prediction of electric vehicle charging load, which provides an important basis for determining the site selection and capacity of electric vehicle charging stations.

[0003] The invention patent with application number 202010013815.9 provides a method for predicting the spatiotemporal distribution of electric vehicle charging load based on the travel probability matrix. First, a probability model of influencing factors is established, and then an electric vehicle travel probability matrix is ​​established based on the network topology of electric vehicles and the travel of electric vehicles between cities. Finally, the Monte Carlo method is used to predict the spatiotemporal distribution of electric vehicle charging load for one day based on the probability model of influencing factors and the electric vehicle travel probability matrix. Although this method can predict the charging load of electric vehicles under spatiotemporal distribution, it does not take into account the spatial perception of electric vehicle users on driving distance and congested road avoidance behavior, and cannot meet the needs of electric vehicle users to quickly reach their destinations in actual driving, so the prediction error is large. Summary of the invention

[0004] The purpose of the present invention is to provide a method for predicting the spatiotemporal distribution of electric vehicle charging load based on information-physical fusion, which meets the actual needs of electric vehicle users to reach the destination quickly and has high prediction accuracy, in order to solve the above problems existing in the prior art.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows:

[0006] The present invention provides a method for predicting the spatiotemporal distribution of electric vehicle charging load based on cyber-physical fusion, the method comprising:

[0007] S1. Obtain the number of electric vehicles in use;

[0008] S2, generating the parking time, initial travel time, initial power, starting position, and end position of a single electric vehicle, using the comprehensive road impedance model to perform the shortest path planning to obtain the shortest travel path from the starting position to the end position of the electric vehicle, using the power consumption model combined with the shortest travel path to calculate the remaining battery power of the electric vehicle in real time, and judging in real time whether the remaining battery power meets the charging condition. When the charging condition is met, the electric vehicle generates a charging demand, and the time when the electric vehicle generates the charging demand is recorded, and then entering S3;

[0009] S3, calculating the maximum drivable distance of the electric vehicle based on the remaining power when the electric vehicle generates a charging demand, determining a set of optional charging nodes based on the maximum drivable distance and the location where the electric vehicle generates a charging demand, randomly selecting a charging node from the set of optional charging nodes as the final charging node, and recording the final charging node;

[0010] S4. Use the comprehensive road impedance model to perform the shortest path planning to obtain the shortest charging path for the electric vehicle from the location where the charging demand is generated to the final charging node. Use the power consumption model combined with the shortest charging path to calculate the remaining battery power of the electric vehicle when it reaches the final charging node from the location where the charging demand is generated, and record the time when the electric vehicle arrives at the final charging node.

[0011] S5, calculating the time required for the electric vehicle to queue for charging after arriving at the final charging node, starting charging after the queue ends, selecting a charging mode based on the parking time of the electric vehicle and the remaining battery power when arriving at the final charging node, and recording the start time of charging and the charging power under the selected charging mode;

[0012] S6. Determine whether the record of the final charging node, the start charging time and the charging power of all electric vehicles has been completed. If not, return to step S2 to calculate the next electric vehicle. If so, output the dynamic prediction result of the spatiotemporal distribution of the electric vehicle charging load.

[0013] The method for obtaining the number of electric vehicles is as follows: electric vehicle penetration rate, electric vehicle range, number of regional public charging piles, per capita GDP, and government subsidies are selected as input data to train a BP neural network to predict the number of electric vehicles.

[0014] The simulated annealing algorithm is used to optimize the weights of the BP neural network. The specific steps are:

[0015] S11, setting the initial temperature and cooling rate of the simulated annealing algorithm; generating an initial solution, that is, randomly generating an initial weight set of the BP neural network;

[0016] S12. Generate a new solution based on the current solution, calculate the fitness difference ΔE between the current solution and the new solution, if ΔE < 0, accept the new solution, if ΔE ≥ 0, decide whether to accept the new solution according to the Metropolis criterion; the Metropolis criterion means that if the randomly generated random number is less than the probability P, the new solution is accepted, and the probability P = e -ΔE / Ts , Ts is the current temperature, otherwise the new solution is not accepted;

[0017] S13, update the current solution; update the temperature according to the following formula: Ts2 = Ts1 × α, where Ts1 and Ts2 represent the temperature before and after the update, respectively, and α m represents the cooling rate;

[0018] S14, return to S12 to perform iterative optimization until the maximum number of iterations is reached, and select the current solution as the optimal weight to train the BP neural network.

[0019] The shortest path planning using the comprehensive road impedance model refers to: obtaining the comprehensive road impedance of each road section in the traffic network through the comprehensive road impedance model, and using the comprehensive road impedance of each road section as the weight of each road section to perform the shortest path planning using the Floyd algorithm.

[0020] The comprehensive road impedance model includes:

[0021]

[0022] In the above formula, R represents the comprehensive road impedance of the road between node i and node j at time t; ij (t) represents the travel time between nodes i and j; C i (t) represents the waiting time for the signal light at node i at time t; It represents the travel time between the roads from node i to node j when the road saturation is 0≤S≤1 at time t; It represents the travel time between nodes i and j at time t when the road saturation is 1<S≤2; It indicates the time to wait for the traffic light at node i when the road saturation is 0<S≤0.6 at time t; Denote the waiting time for the signal light at node \(i\) when the road saturation \(S>0.6\) at time \(t\); \(t_0\) represents the passing time required for an electric vehicle to pass through the road when the road has zero traffic flow; \(\alpha\) and \(\beta\) are both influencing factors; \(S\) is the road saturation, \(0 < S\leq0.6\) indicates that the road is unobstructed, \(0.6 < S\leq0.8\) indicates that the road is moving slowly, \(0.8 < S\leq1\) indicates that the road is congested, \(1 < S\leq2\) indicates that the road is severely congested, \(S = Q / C\), where \(Q\) is the road traffic flow and \(C\) is the road capacity, that is, the number of vehicles that can pass smoothly through the road per unit time; \(c\) is the signal cycle, which refers to the time required for the traffic signal to go through one cycle of red - green - yellow; \(\lambda\) is the ratio of the green - light signal time length to the total signal - light time length; \(q\) represents the vehicle arrival rate, that is, the number of electric vehicles arriving at node \(i\) at time \(t\).

[0023] The power consumption model includes:

[0024]

[0025] In the above formula, is the remaining power; is the initial power; \(C\) t is the battery capacity of the electric vehicle; \(d\) ij is the road length between node \(i\) and node \(j\); \(E\) p is the energy consumption per unit mileage of the electric vehicle; is the power consumption of the air - conditioner when the electric vehicle is traveling at speed \(V\) [t,t+Δt] , traveling a distance of \(d\) [t，t+Δt] and the environmental temperature is \(T\); is the power consumption generated when the electric vehicle travels on the road at speed \(V\) [t,t+Δt] ; are respectively the power consumptions generated when the electric vehicle travels at speed \(V\) [t，t+Δt] on the main road and the secondary road; are respectively the cooling and heating powers of the air - conditioner; \(T\) k-max , \(T\) k-min are respectively the cold and hot limit temperatures; \(V\) [t,t+Δt] represents the driving speed of the electric vehicle in the time period \([t,t + \Delta t]\); \(V\) s is the designed road speed; \(C\) is the road capacity, that is, the number of vehicles that can pass smoothly through the road per unit time; \(N\) t is the number of vehicles traveling on the road in the time period \([t,t+\Delta t]\); \(\gamma\) is a non - linear function; \(\lambda_1\), \(\lambda_2\), \(\lambda_3\) are all adaptive parameters.

[0026] Calculate the waiting time required for the electric vehicle to queue up for charging after reaching the final charging node according to the following formula:

[0027]

[0028] In the above formula, t w is the waiting time of electric vehicles; c is the number of charging piles in the charging node; μ is the number of electric vehicles that can be served by each charging pile per unit time; ρ is the service intensity of the charging pile; P0 is the probability of electric vehicles accepting charging services; k represents the kth charging pile; k! represents the factorial of k; c! represents the factorial of c; L s is the average captain; ω is the number of charging demands for services at the charging station per hour.

[0029] The method of selecting a charging mode based on the parking time of the electric vehicle and the remaining battery power when the electric vehicle reaches the final charging node is as follows: the time required for the electric vehicle to be charged to the desired charging level in the slow charging mode is calculated according to the following formula, and it is determined whether the time required for the electric vehicle to be charged to the desired charging level in the slow charging mode is greater than the parking time. If so, the electric vehicle is charged in the fast charging mode, otherwise, the electric vehicle is charged in the slow charging mode;

[0030]

[0031] In the above formula, t sc The time required to charge an electric vehicle to the desired charge level in slow charging mode; is the expected power; B is the remaining power of the electric vehicle when it reaches the final charging node; c is the battery capacity of the electric vehicle; P sc is the charging power of the electric vehicle in slow charging mode; η is the charging efficiency of the electric vehicle.

[0032] The charging condition refers to the remaining battery power reaching a charging threshold.

[0033] The expression of the spatiotemporal distribution of the electric vehicle charging load is:

[0034]

[0035] In the above formula, P l (t) is the charging load at charging node l at time t; is the charging power of the i-th electric vehicle connected to the charging node l at time t under the selected charging mode.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. The method for predicting the spatiotemporal distribution of electric vehicle charging load based on information-physical fusion described in the present invention generates the parking time, initial travel time, initial power, starting position, and end position of a single electric vehicle, uses a comprehensive road impedance model to perform shortest path planning to obtain the shortest travel path of the electric vehicle from the starting position to the end position, uses a power consumption model combined with the shortest travel path to calculate the remaining battery power of the electric vehicle in real time, and judges in real time whether the remaining battery power meets the charging condition. When the charging condition is met, the electric vehicle generates a charging demand, and the time when the electric vehicle generates the charging demand is recorded. Subsequently, based on the remaining power when the electric vehicle generates the charging demand, the maximum drivable distance of the electric vehicle is calculated, and based on the maximum drivable distance and the location where the electric vehicle generates the charging demand, a set of optional charging nodes is determined, and a charging node is randomly selected from the set of optional charging nodes as the final charging node, and the final charging node is recorded. The shortest path planning is performed using a comprehensive road impedance model to obtain the shortest charging path for the electric vehicle from the location where the charging demand is generated to the final charging node, and the power consumption model is used in combination with the shortest charging path to calculate the distance from the location where the charging demand is generated to the final charging node. The remaining battery power when the electric vehicle reaches the final charging node is set, the time when the electric vehicle reaches the final charging node is recorded, the time required for the electric vehicle to queue for charging after reaching the final charging node is calculated, and charging starts after the queue ends. The charging mode is selected based on the parking time of the electric vehicle and the remaining battery power when the electric vehicle reaches the final charging node, and the charging start time and the charging power under the selected charging mode are recorded. After completing the recording of all the final charging nodes, the charging start time and the charging power of all electric vehicles, the dynamic prediction result of the spatiotemporal distribution of the charging load of the electric vehicle is output; on the one hand, the above design performs the shortest path planning for the driving path of the electric vehicle to the terminal position and the driving path of the electric vehicle to the charging node after the charging demand is generated, which meets the user's perception of the shortest driving distance, and can meet the needs of electric vehicle users to quickly reach the destination in actual driving behavior, so that the charging node can be determined more accurately in the future, thereby more accurately describing the spatiotemporal distribution of the electric vehicle. On the other hand, the influence of the congestion of the charging station on the spatiotemporal distribution of the electric vehicle is described by calculating the waiting time in the queue, and finally the prediction accuracy of the dynamic prediction result of the spatiotemporal distribution of the charging load of the electric vehicle is improved. Therefore, the present invention meets the needs of electric vehicle users to quickly reach the destination in actual driving behavior and can improve the prediction accuracy.

[0038] 2. The method for predicting the spatiotemporal distribution of electric vehicle charging load based on information-physical fusion described in the present invention is combined with data-driven on the basis of model-driven, that is, the BP neural network optimized by simulated annealing algorithm is first used to predict the number of electric vehicles, and then the spatiotemporal distribution of charging load is dynamically predicted based on the predicted number of electric vehicles, which is suitable for application in areas where large-scale access of electric vehicles is planned. Therefore, the present invention is suitable for application in areas where large-scale access of electric vehicles is planned.

[0039] 3. The method for predicting the spatiotemporal distribution of electric vehicle charging load based on information-physical fusion described in the present invention not only integrates the road section impedance and the node impedance caused by the time delay generated under the traffic light control in the constructed comprehensive road impedance model, but also further divides the impedance according to the road saturation, which is more in line with the user's spatial perception of the congested road ahead in actual driving behavior, thereby making the planning results of the shortest path planning more accurate, and further improving the accuracy of the prediction results of the spatiotemporal distribution of charging load. Therefore, the present invention is in line with the user's spatial perception of the congested road ahead in actual driving behavior, and can further improve the prediction accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 The present invention is a flowchart of the method for predicting the spatiotemporal distribution of charging load. DETAILED DESCRIPTION

[0041] The present invention is further described in detail below in conjunction with specific implementations and drawings.

[0042] Example:

[0043] See also Figure 1 , a method for predicting the spatiotemporal distribution of electric vehicle charging load based on cyber-physical fusion is carried out in the following steps:

[0044] S1. Obtain the number of electric vehicles in use. The method is as follows: select the penetration rate of electric vehicles, the range of electric vehicles, the number of regional public charging piles, the per capita GDP, and the government subsidy as input data to train the BP neural network and predict the number of electric vehicles in use. Since the initial weights of the BP neural network are not optimized, it is easy to fall into the local minimum and converge slowly. Therefore, the weights of the BP neural network are optimized using the simulated annealing algorithm. The specific steps are as follows:

[0045] S11, setting the initial temperature and cooling rate of the simulated annealing algorithm, the initial temperature is used to determine the randomness of the search, the cooling rate is used to decay the temperature of the next round according to the current temperature, and the cooling rate is generally set to 0.95; and generating an initial solution, that is, randomly generating an initial weight set of the BP neural network;

[0046] S12, generating a new solution based on the current solution;

[0047] S13, calculate the fitness difference ΔE between the current solution and the new solution, if ΔE < 0, accept the new solution, if ΔE ≥ 0, decide whether to accept the new solution according to the Metropo]is criterion; the Metropo]is criterion means that if the randomly generated random number is less than the probability P, the new solution is accepted, and the probability P = e -ΔΔE / Ts , Ts is the current temperature, otherwise the new solution is not accepted;

[0048] S14, updating the current solution;

[0049] S15. Take S12-S14 as an iterative optimization process. After each iterative optimization, update the temperature according to the following formula: Ts2 = Ts1 × α, where Ts1 and Ts2 represent the temperature before and after the update, respectively, and α m Indicates the cooling rate; the updated temperature is used as the current temperature;

[0050] S16, return to S12 to perform the next iterative optimization until the maximum number of iterations is reached;

[0051] S17, selecting the current solution as the optimal weight to train the BP neural network;

[0052] S2. Read the type of a single electric vehicle, and generate basic parameters of the electric vehicle based on the type of the electric vehicle. The basic parameters of the electric vehicle include parking time, initial travel time, battery capacity, initial power, etc. The initial travel time refers to the time when the electric vehicle sets out from the starting position for the first time in a day. For example, electric vehicles include three types: taxis, private cars, and urban function vehicles. The parking time, initial travel time, battery capacity, and initial power of different types of electric vehicles are different, which can be obtained by statistically analyzing the probability distribution of the basic parameters corresponding to the electric vehicle type; the terminal position of the electric vehicle is obtained based on the starting position of the electric vehicle and the travel probability matrix;

[0053] The shortest path planning is performed using the comprehensive road impedance model to obtain the shortest travel path of the electric vehicle from the starting position to the end position; the remaining battery power of the electric vehicle is calculated in real time using the power consumption model combined with the shortest travel path, and it is determined in real time whether the remaining battery power reaches the charging condition, the charging condition refers to the remaining battery power reaching the charging threshold, and the charging threshold is generally set to 20% of the battery capacity; when the charging threshold is reached, the electric vehicle generates a charging demand, and the time when the electric vehicle generates the charging demand is recorded, and enters S3;

[0054] The shortest path planning using the comprehensive road impedance model means: obtaining the comprehensive road impedance of each section in the traffic network through the comprehensive road impedance model, and using the Floyd algorithm for shortest path planning with the comprehensive road impedance of each section as the weight of each section; since when an electric vehicle passes through a road, in addition to being affected by the road resistance itself, it is also affected by the traffic light control at the road node, and more time delays will occur under the signal control of the traffic light; therefore, when designing the comprehensive road impedance model of the present invention, the road impedance and the node impedance caused by the time delay generated under the traffic light control are fused; the comprehensive road impedance model includes:

[0055]

[0056] In the above formula, represents the comprehensive road impedance of the road between node i and node j at time t; R ij (t) represents the travel time between the road between node i and node j; C i (t) represents the waiting time for the signal light at node i at time t; represents the travel time between the road between node i and node j at time t when the road saturation degree 0≤S≤1; represents the travel time between the road between node i and node j at time t when the road saturation degree 1<S≤2; represents the waiting time for the signal light at node i at time t when the road saturation degree 0<S≤0.6; represents the waiting time for the signal light at node i at time t when the road saturation degree S>0.6; t0 represents the travel time required for the electric vehicle to pass through the road when the road has zero traffic flow; α and β are both influencing factors; S is the road saturation degree, 0<S≤0.6 means the road is unobstructed, 0.6<S≤0.8 means the road is slowly passing, 0.8<S≤1 means the road is congested, 1<S≤2 means the road is severely congested, S = Q / C, Q is the road traffic flow, C is the road capacity, that is, the number of vehicles that can pass smoothly on the road per unit time; c is the signal cycle, which refers to the time required for the traffic signal to go through a cycle of red - green - yellow; λ is the ratio of the green light signal time length to the total signal light time length; q represents the vehicle arrival rate, that is, the number of electric vehicles arriving at node i at time t;

[0057] When designing the power consumption model of the present invention, the influence of road capacity and environmental temperature on power consumption is considered, so as to more accurately describe the user's driving habits in complex road conditions; the power consumption model includes:

[0058]

[0059]

[0060]

[0061] In the above formula, is the remaining power; is the initial charge; C t is the battery capacity of the electric vehicle; d ij is the length of the road between node i and node j; E p is the energy consumption per unit mileage of electric vehicles; The electric vehicle is traveling at a speed of V [t,t+Δt] , the driving distance is d [t,t+Δt] , air conditioning power consumption when the ambient temperature is T; The electric vehicle is traveling at a speed of V [t,t+Δt] The amount of electricity consumed while driving on the road; The electric vehicle is driven at a speed of V [t,t+Δt] The power consumption generated by driving on main roads and secondary roads; are the cooling and heating power of air conditioning respectively; T k-max , T k-min are the cold and hot limit temperatures respectively; V [t,t+Δt] represents the driving speed of the electric vehicle in the period [t, t+Δt]; V s is the design speed of the road; C is the road capacity, that is, the number of vehicles that can pass smoothly on the road in unit time; N t is the number of vehicles traveling on the road during the period [t, t+Δt]; γ is a nonlinear function; λ1, λ2, and λ3 are all adaptive parameters;

[0062] S3, calculating the maximum drivable distance of the electric vehicle based on the remaining power of the electric vehicle when the charging demand is generated, determining a set of optional charging nodes based on the maximum drivable distance and the location of the electric vehicle when the charging demand is generated, randomly selecting a charging node from the set of optional charging nodes as the final charging node, and recording the final charging node;

[0063] S4. Use the comprehensive road impedance model to perform the shortest path planning to obtain the shortest charging path for the electric vehicle from the location where the charging demand is generated to the final charging node. Use the power consumption model combined with the shortest charging path to calculate the remaining battery power of the electric vehicle when it reaches the final charging node from the location where the charging demand is generated, and record the time when the electric vehicle arrives at the final charging node.

[0064] S5. To accurately describe the congestion of charging stations, the occupancy rate of charging piles at the final charging node in each time period is obtained by statistics. If the occupancy rate of charging piles is less than or equal to 1 when the electric vehicle arrives at the final charging node, the electric vehicle can be charged directly without waiting. At this time, the waiting time t wIf the charging pile occupancy rate is greater than 1, the time required for electric vehicles to queue up and wait for charging after arriving at the final charging node is calculated according to the following formula:

[0065]

[0066]

[0067]

[0068]

[0069] In the above formula, t w is the time it takes for electric vehicles to queue for charging; c is the number of charging piles in the charging node; μ is the number of electric vehicles that can be served by each charging pile per unit time; ρ is the service intensity of the charging pile; P0 is the probability of an electric vehicle accepting charging service; k represents the kth charging pile; k! represents the factorial of k; c! represents the factorial of c; L s is the average captain; ω is the number of charging demands received at the charging station per hour;

[0070] After the queue ends, charging begins. First, the charging mode is selected based on the parking time of the electric vehicle and the remaining battery power when it reaches the final charging node. Then, the time required for the electric vehicle to charge to the expected charging level under the selected charging mode is calculated. The charging start time, the charging power and the charging time under the selected charging mode are recorded. The charging time of the electric vehicle shall not exceed the parking time.

[0071] The charging mode is selected based on the parking time of the electric vehicle and the remaining battery power when the electric vehicle reaches the final charging node. Specifically, the time required for the electric vehicle to be charged to the desired charging level in the slow charging mode is calculated according to the following formula, and it is determined whether it is greater than the parking time. If so, the electric vehicle is charged in the fast charging mode, otherwise, the electric vehicle is charged in the slow charging mode.

[0072]

[0073] In the above formula, t sc The time required to charge an electric vehicle to the desired charge level in slow charging mode; is the expected power; B is the remaining power of the electric vehicle when it reaches the final charging node; c is the battery capacity of the electric vehicle; P sc is the charging power of the electric vehicle in slow charging mode; η is the charging efficiency of the electric vehicle;

[0074] The calculation formula for the time required for an electric vehicle to charge to the desired charge level in slow charging mode is:

[0075]

[0076] In the above formula, t fc The time required to charge an electric vehicle to the desired charge level in fast charging mode; fc The charging power of electric vehicles in fast charging mode;

[0077] S6: Determine whether the record of the final charging node, charging start time, charging power, and charging duration of all electric vehicles has been completed. If not, return to step S2 to calculate the next electric vehicle. If the calculation of the charging power of all electric vehicles has been completed, output the dynamic prediction result of the spatiotemporal distribution of the electric vehicle charging load; the expression of the spatiotemporal distribution of the electric vehicle charging load is:

[0078]

[0079] In the above formula, P l (t) is the charging load at charging node l at time t; is the charging power of the i-th electric vehicle connected to the charging node l at time t under the selected charging mode.

Claims

1. A method for predicting the spatiotemporal distribution of electric vehicle charging load based on cyber-physical fusion, characterized by: The method comprises: S1. Obtain the number of electric vehicles in use; S2, generating the parking time, initial travel time, initial power, starting position, and end position of a single electric vehicle, using the comprehensive road impedance model to perform the shortest path planning to obtain the shortest travel path from the starting position to the end position of the electric vehicle, using the power consumption model combined with the shortest travel path to calculate the remaining battery power of the electric vehicle in real time, and judging in real time whether the remaining battery power meets the charging condition. When the charging condition is met, the electric vehicle generates a charging demand, and the time when the electric vehicle generates the charging demand is recorded, and then entering S3; S3, calculating the maximum drivable distance of the electric vehicle based on the remaining power when the electric vehicle generates a charging demand, determining a set of optional charging nodes based on the maximum drivable distance and the location where the electric vehicle generates a charging demand, randomly selecting a charging node from the set of optional charging nodes as the final charging node, and recording the final charging node; S4. Use the comprehensive road impedance model to perform the shortest path planning to obtain the shortest charging path for the electric vehicle from the location where the charging demand is generated to the final charging node. Use the power consumption model combined with the shortest charging path to calculate the remaining battery power of the electric vehicle when it reaches the final charging node from the location where the charging demand is generated, and record the time when the electric vehicle arrives at the final charging node. S5, calculating the time required for the electric vehicle to queue for charging after arriving at the final charging node, starting charging after the queue ends, selecting a charging mode based on the parking time of the electric vehicle and the remaining battery power when arriving at the final charging node, and recording the start time of charging and the charging power under the selected charging mode; S6. Determine whether the record of the final charging node, the start charging time and the charging power of all electric vehicles has been completed. If not, return to step S2 to calculate the next electric vehicle. If so, output the dynamic prediction result of the spatiotemporal distribution of the electric vehicle charging load.

2. The method for predicting the spatiotemporal distribution of electric vehicle charging load based on cyber-physical fusion according to claim 1 is characterized in that: The method for obtaining the number of electric vehicles is as follows: electric vehicle penetration rate, electric vehicle range, number of regional public charging piles, per capita GDP, and government subsidies are selected as input data to train a BP neural network to predict the number of electric vehicles.

3. The method for predicting the spatiotemporal distribution of electric vehicle charging load based on cyber-physical fusion according to claim 2 is characterized by: The simulated annealing algorithm is used to optimize the weights of the BP neural network. The specific steps are: S11, setting the initial temperature and cooling rate of the simulated annealing algorithm; generating an initial solution, that is, randomly generating an initial weight set of the BP neural network; S12. Generate a new solution based on the current solution, calculate the fitness difference ΔE between the current solution and the new solution, if ΔE < 0, accept the new solution, if ΔE ≥ 0, decide whether to accept the new solution according to the Metropolis criterion; the Metropolis criterion means that if the randomly generated random number is less than the probability P, the new solution is accepted, and the probability P = e -ΔE / Ts , Ts is the current temperature, otherwise the new solution is not accepted; S13, updating the current solution; The temperature is updated according to the following formula: Ts2 = Ts1 × α, where Ts1 and Ts2 represent the temperature before and after the update, respectively, and α m represents the cooling rate; S14, return to S12 to perform iterative optimization until the maximum number of iterations is reached, and select the current solution as the optimal weight to train the BP neural network.

4. The method for predicting the spatiotemporal distribution of electric vehicle charging load based on cyber-physical fusion according to any one of claims 1 to 3, characterized in that: The shortest path planning using the comprehensive road impedance model refers to: obtaining the comprehensive road impedance of each road section in the traffic network through the comprehensive road impedance model, and using the comprehensive road impedance of each road section as the weight of each road section to perform the shortest path planning using the Floyd algorithm.

5. The method for predicting the spatiotemporal distribution of electric vehicle charging load based on cyber-physical fusion according to any one of claims 1 to 3, characterized in that: The comprehensive road impedance model includes: In the above formula, represents the comprehensive road impedance of the road between node i and node j at time t; R ij (t) represents the travel time between the roads between node i and node j; C i (t) represents the waiting time for the signal light at node i at time t; represents the travel time between the roads between node i and node j at time t when the road saturation is 0 ≤ S ≤ 1; represents the travel time between the roads between node i and node j at time t when the road saturation is 1 < S ≤ 2; represents the waiting time for the signal light at node i when the road saturation is 0 < S ≤ 0.6 at time t; represents the waiting time for the signal light at node i when the road saturation is S > 0.6 at time t; t0 represents the travel time required for the electric vehicle to pass through the road when the road has zero traffic flow; α and β are both influencing factors; S is the road saturation, 0 < S ≤ 0.6 indicates that the road is unobstructed, 0.6 < S ≤ 0.8 indicates that the road is moving slowly, 0.8 < S ≤ 1 indicates that the road is congested, 1 < S ≤ 2 indicates that the road is severely congested, S = Q / C, Q is the road traffic flow, C is the road capacity, that is, the number of vehicles that can pass smoothly through the road per unit time; c is the signal cycle, which refers to the time required for the traffic signal to go through a red-green-yellow cycle; λ is the ratio of the green signal time length to the total signal light time length; q represents the vehicle arrival rate, that is, the number of electric vehicles arriving at node i at time t.

6. The method for predicting the spatiotemporal distribution of electric vehicle charging load based on cyber-physical fusion according to any one of claims 1 to 3, characterized in that: The power consumption model includes: In the above formula, is the remaining power; is the initial charge; C t is the battery capacity of the electric vehicle; d ij is the length of the road between node i and node j; E p is the energy consumption per unit mileage of electric vehicles; The electric vehicle is traveling at a speed of V [t,t+Δt] , the driving distance is d [t,t+Δt] , air conditioning power consumption when the ambient temperature is T; The electric vehicle is traveling at a speed of V [t,t+Δt] The amount of electricity consumed while driving on the road; The electric vehicle is driven at a speed of V [t,t+Δt] The power consumption generated by driving on main roads and secondary roads; are the cooling and heating power of air conditioning respectively; T k-max , T k-min are the cold and hot limit temperatures respectively; V [t,t+Δt ] represents the driving speed of the electric vehicle in the period [t, t+Δt]; V s is the design speed of the road; C is the road capacity, that is, the number of vehicles that can pass smoothly on the road in unit time; N t is the number of vehicles traveling on the road during the period [t, t+Δt]; γ is a nonlinear function; λ1, λ2, and λ3 are all adaptive parameters.

7. The method for predicting the spatiotemporal distribution of electric vehicle charging load based on cyber-physical fusion according to claim 6 is characterized by: The time required for electric vehicles to wait in line for charging after arriving at the final charging node is calculated according to the following formula: In the above formula, t w The waiting time in the queue for electric vehicles; c is the number of charging piles in the charging node; μ is the number of electric vehicles that can be served by each charging pile per unit time; ρ is the service intensity of the charging pile; P0 is the probability of an electric vehicle accepting charging service; k represents the kth charging pile; k! represents the factorial of k; c! represents the factorial of c; L s is the average captain; ω is the number of charging demands for services at the charging station per hour.

8. The method for predicting the spatiotemporal distribution of electric vehicle charging load based on cyber-physical fusion according to claim 7 is characterized by: The method of selecting a charging mode based on the parking time of the electric vehicle and the remaining battery power when the electric vehicle reaches the final charging node is as follows: the time required for the electric vehicle to be charged to the desired charging level in the slow charging mode is calculated according to the following formula, and it is determined whether the time required for the electric vehicle to be charged to the desired charging level in the slow charging mode is greater than the parking time. If so, the electric vehicle is charged in the fast charging mode, otherwise, the electric vehicle is charged in the slow charging mode; In the above formula, t sc The time required to charge an electric vehicle to the desired charge level in slow charging mode; is the expected power; B is the remaining power of the electric vehicle when it reaches the final charging node; c is the battery capacity of the electric vehicle; P sc is the charging power of the electric vehicle in slow charging mode; η is the charging efficiency of the electric vehicle.

9. The method for predicting the spatiotemporal distribution of electric vehicle charging load based on cyber-physical fusion according to any one of claims 1 to 3, characterized in that: The charging condition refers to the remaining battery power reaching a charging threshold.

10. The method for predicting the spatiotemporal distribution of electric vehicle charging load based on cyber-physical fusion according to any one of claims 1 to 3, characterized in that: The expression of the spatiotemporal distribution of the electric vehicle charging load is: In the above formula, P l (t) is the charging load at charging node l at time t; is the charging power of the i-th electric vehicle connected to the charging node l at time t under the selected charging mode.

Images