Electric vehicle charging scheduling method based on improved optimization algorithm and terminal

Abstract

The invention discloses an electric vehicle charging scheduling method based on an improved optimization algorithm and a terminal. The method comprises the steps that a user side terminal system needs to judge whether a user sends a charging request or not according to the state of charge of an electric vehicle battery, if yes, a data acquisition unit obtains the daily charging demand, position information, power distribution network operation conditions, regional charging station information and related data information of a regional traffic network of the electric vehicle, and if not, the user side terminal system sends the charging request; if yes, returning to the previous step for cyclic judgment; the data processing and algorithm calculation unit establishes an optimization model with the minimum power grid load fluctuation and the minimum economic time cost of user charging as targets, inputs the obtained data into the model, and solves a target function by adopting a non-dominated sorting whale optimization algorithm with an elitist strategy; the optimal charging scheme of the electric vehicle is obtained; and the transmission and application unit sends charging scheme suggestions to users, so that ordered charging scheduling of the electric vehicles in the region is realized.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent charging optimization scheduling of electric vehicles, and in particular, relates to an electric vehicle charging scheduling method and terminal based on an improved optimization algorithm. Background Art

[0002] As environmental pollution, global warming and fossil energy shortages become increasingly apparent, electric vehicles are favored as a green travel mode. With the continuous introduction of national policies to support electric vehicles, the number of new energy vehicles has shown a fluctuating upward trend. At the same time, the charging load of electric vehicles has also increased. A large amount of disordered load demand will bring a huge impact to the power grid, affecting the smooth operation of the power system, increasing the peak-to-valley difference of the power grid load, and reducing the quality of power.

[0003] Assuming that electric vehicles have a demand for charging during driving, if the user-side system can generate an intelligent and orderly charging plan and provide relevant charging suggestions based on the location of the electric vehicle, relevant information of nearby charging stations, traffic network and other data, then the adverse effects of a large number of electric vehicle loads connected to the power grid can be effectively avoided. The electric vehicle charging scheduling method and terminal based on the improved optimization algorithm provide an accurate and reliable solution, which can effectively select a charging strategy with lower charging cost and less impact on the stability of the power system for electric vehicles with charging needs, thereby promoting the development of the electric vehicle industry. Summary of the invention

[0004] In order to overcome the adverse effects of disorderly charging of a large number of electric vehicles and provide users with a more economical and time-saving charging solution, the present invention provides an electric vehicle charging scheduling method and terminal based on an improved optimization algorithm.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] An electric vehicle charging scheduling terminal includes a data acquisition unit, a data processing and algorithm calculation unit, and a transmission and application unit. During the process of electric vehicle charging scheduling, the terminal tracks changes in the state of charge of the electric vehicle battery, makes charging request decisions, and sends corresponding charging suggestions to users.

[0007] A charging scheduling method for electric vehicles based on an improved optimization algorithm, the scheduling method comprising:

[0008] S1. The electric vehicle charging dispatch terminal needs to determine whether the user has issued a charging request based on the charge state of the electric vehicle battery on the user side. If so, proceed to the next step. If not, return to the previous step for loop judgment;

[0009] S2. The data acquisition unit obtains the daily charging demand and location of electric vehicles, information on charging stations accessible to electric vehicles, regional transportation networks, and operation data of regional distribution networks;

[0010] S3. Input the acquired relevant data information into the data processing and algorithm calculation unit of the terminal, filter, analyze, extract, model and solve the data through the data processing and algorithm calculation unit to obtain the optimal charging strategy for the electric vehicle;

[0011] S4. The transmission and application unit of the terminal sends charging plan suggestions to the user to achieve orderly charging scheduling of electric vehicles in the region;

[0012] Furthermore, the data acquisition unit obtains the daily charging demand of the electric vehicle, the location of the electric vehicle, including the charging demand of the electric vehicle within a day, the battery charge state when the electric vehicle issues a charging demand, and the location and time information; the information of the charging stations reachable by the electric vehicle includes the location of the charging stations reachable near the place, the charging power and efficiency of the charging stations, and the time-sharing charging electricity price; the distribution network operation data of the region includes the basic load data of the regional distribution network operation.

[0013] Furthermore, the data acquisition unit obtains the daily charging demand and location of the electric vehicle, the information of the charging stations accessible to the electric vehicle, the regional traffic network, and the operation data of the regional distribution network, and inputs them into the data processing and algorithm calculation unit of the terminal, which specifically includes the following steps:

[0014] S31. With the goal of minimizing grid load fluctuation, minimizing the time cost of electric vehicle users charging, and minimizing the economic cost, the charging status of electric vehicle users is taken as the decision variable, and the charging efficiency of the charging station, the charging capacity of the electric vehicle, and the value range of the decision variable are taken as the constraints, a multi-objective optimization model is established;

[0015] S32. Use the non-dominated sorting whale optimization algorithm with an elite strategy to solve the multi-objective optimization model of the electric vehicle charging scheduling to obtain the optimal charging strategy.

[0016] Furthermore, S 31 Specifically, define the decision variable x i,j,k , minimize the grid load fluctuation F1(x i,j,k ), user charging economic cost F2(x i,j,k ), user charging time cost F3(x i,j,k ), the mathematical expressions of the objective function and constraints are as follows:

[0017]

[0018]

[0019] Among them, F1(x i,j,k ) is the grid load fluctuation function, F2(x i,j,k ) is the user charging economic cost function, F3(x i,j,k ) is the user charging time cost function, B i,j,k represents the load fluctuation rate caused by electric vehicle i charging at charging station j during time period k, P i,j,k represents the peak-to-valley difference of the load when electric vehicle i is charging at charging station j during time period k, V i,j,k represents the load variance of electric vehicle i charging at charging station j during time period k, a, b, c are the index weight coefficients, P max represents the maximum value of the load fluctuation curve of electric vehicle i after charging at charging station j, P av Indicates the average value of the load fluctuation curve, P min Indicates the minimum value of the load fluctuation curve, S j represents the service fee of charging station j, pr k represents the time-of-use electricity price in period k, t i,j,k represents the charging time of electric vehicle i at charging station j during time period k, L i,j represents the path length from electric vehicle i to charging station j, v i Indicates the average driving speed, tp j,k represents the queuing time required by charging station j in time period k, ΔC i Indicates the required charging capacity, η j represents the charging efficiency of charging station j, P j represents the charging power provided by charging station j.

[0020] Furthermore, S 32 The specific steps include:

[0021] Step 1): Initialize the optional charging station locations for electric vehicles, calculate the fitness of the optional charging stations, and obtain the current optimal location through non-dominated sorting And record, set the maximum number of iterations, and set the current number of iterations t to 1;

[0022] Step 2): Generate random numbers p, r1, r2 in [0,1], determine whether p>0.5, if so, proceed to step 3), otherwise proceed to step 4), set parameter a to decrease linearly in the range of [0,2], and calculate coefficient vectors A and C;

[0023] A=2arQ

[0024] C2r2

[0025] Step 3): Generate a random number in [-1,1] and update the position according to the following spiral search formula;

[0026]

[0027] Step 4): Determine whether |A|<1. If so, adopt the encircling search mechanism, otherwise perform a random search in the global scope. The formulas for encircling search and random search are:

[0028]

[0029] Step 5): After step 2) to step 4), a new generation of updated solutions will be generated. The new solutions will be merged with the initial solutions for non-dominated sorting, and the Pareto set will be constructed to obtain the first generation optimal solution.

[0030] Step 6): Increase the number of iterations, use the optimal solution obtained in the previous generation as the parent of this iteration, repeat steps 2) to 4) to obtain the child generation;

[0031] Step 7): Adopt the elite retention strategy, merge the parent generation and the child generation and perform fast non-dominated sorting, construct the Pareto set, calculate the congestion, obtain the optimal solution of the new generation, and compare the fitness with the optimal solution of the previous generation. If it is better, replace the current optimal charging station location;

[0032] Step 8): After reaching the maximum number of iterations, the program ends and the global optimal solution is obtained. Otherwise, repeat steps 6) to 7).

[0033] The electric vehicle charging scheduling method and terminal based on the improved optimization algorithm provided by the present invention have the following advantages:

[0034] 1) Grid load balancing: Orderly charging scheduling can reasonably allocate the charging time and charging power of electric vehicles, effectively reduce the peak-to-valley difference of the grid, and improve the stability of the grid.

[0035] 2) Improve resource utilization: Sending charging suggestions to users through the terminal can make more efficient use of charging pile resources, avoid resource waste, and improve charging efficiency.

[0036] 3) Improve the user experience of electric vehicle users: The terminal can provide users with more stable and convenient charging services, reduce user queues and waiting time, and improve user satisfaction.

[0037] 4) Improve urban road conditions: Improve urban traffic conditions. Orderly charging can avoid traffic congestion and improve road traffic efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the optimization scheme in the embodiment of the present invention, the following briefly introduces the drawings required for use in the prior art and the embodiments. The following drawings are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0039] Figure 1 A flow chart of an electric vehicle charging scheduling method and terminal based on an improved optimization algorithm provided by the present invention.

[0040] Figure 2 A flowchart of the non-dominated sorting whale optimization algorithm with an elite strategy provided by the present invention.

[0041] Figure 3 The load curves of orderly charging and disorderly charging provided by the present invention.

[0042] Figure 4 The non-dominated sorting whale optimization algorithm with elite strategy provided by the present invention is solved with the traditional non-dominated sorting whale optimization algorithm, and the obtained charging load curves are compared.

[0043] Figure 5 The charging load curves obtained by solving the non-dominated sorting whale optimization algorithm with elite strategy provided by the present invention and NSGA-Ⅱ are compared. DETAILED DESCRIPTION

[0044] To make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application.

[0046] In order to optimize the best solution for electric vehicle intelligent charging, the user-side terminal system needs to determine whether the user has issued a charging request based on the charge state of the electric vehicle battery;

[0047] Specifically, if a charging request is issued by the user side, data information related to the daily operation of the electric vehicle, distribution network, and charging station will be collected.

[0048] Specifically, the daily operation-related data information of the electric vehicle, distribution network and charging station includes the charging demand of the electric vehicle in a day, the battery charge state and the location and time information when the electric vehicle issues the charging demand, the location, charging power and efficiency of the charging stations accessible nearby, the time-of-use charging electricity price in the area, the traffic road network map of the area, and the basic load data of the daily operation of the distribution network.

[0049] Specifically, the above data information is input into the data processing and algorithm calculation unit of the terminal, the data is analyzed and processed, a mathematical model is established, and a solution is obtained through an algorithm;

[0050] Specifically, a mathematical optimization model is established with the goal of minimizing grid load fluctuation, minimizing user economic cost, and minimizing user charging time cost. Whether user i goes to charging station j to charge in the kth time period xi,j,k is used as the decision variable to construct a multi-objective optimization model. The mathematical expression is as follows:

[0051]

[0052] Among them, F1(x i,j,k ) is the grid load fluctuation function, F2(x i,j,k ) is the user charging economic cost function, F3(x i,j,k ) is the user charging time cost function, B i,j,k represents the load fluctuation rate caused by electric vehicle i charging at charging station j during time period k, P i,j,k represents the peak-to-valley difference of the load when electric vehicle i is charging at charging station j during time period k, V i,j,k represents the load variance of electric vehicle i charging at charging station j during time period k, a, b, c are the index weight coefficients, P max represents the maximum value of the load fluctuation curve of electric vehicle i after charging at charging station j, P av Indicates the average value of the load fluctuation curve, P min Indicates the minimum value of the load fluctuation curve, S j represents the service fee of charging station j, pr k represents the time-of-use electricity price in period k, t i,j,k represents the charging time of electric vehicle i at charging station j during time period k, L i,j represents the path length from electric vehicle i to charging station j, v i Indicates the average driving speed, tp j,k represents the queuing time required by charging station j in time period k, ΔC i Indicates the required charging capacity, η j represents the charging efficiency of charging station j, P j represents the charging power provided by charging station j.

[0053] Specifically, according to the above optimization objective function and constraints, the non-dominated sorting whale optimization algorithm with elite strategy is substituted to optimize the optimal charging scheduling plan for a certain area in the following steps:

[0054] Step 1): Initialize the position vectors of all N optional charging stations for electric vehicles Calculate the grid load fluctuation, user time cost and economic cost of the currently selected charging station location, and find the current optimal solution F through non-dominated sorting ini (x i,j,k ), and record the location of the optimal charging station at this time as Set the maximum number of iterations to t max , set the current number of iterations t = 1;

[0055] Step 2): Set the parameter value. First, generate a random number p in [0,1]. If p>0.5, proceed to step 3). Otherwise, jump to step 4). Set a=2. During the iteration, a decreases linearly in the range of [0,2]. Generate random numbers r1 and r2 in [0,1]. Calculate the coefficient vectors A and C according to the following formula.

[0056] A=2ar1-a

[0057] C=2r2

[0058] Step 3): Local search mechanism, using a spiral mechanism to update the position and generate a new position vector. The formula is as follows, where l is a random number in [-1,1];

[0059]

[0060] Step 4): Determine whether |A| is less than 1. If so, update the position through a surround search, otherwise update the position through a random search. The formulas for the surround search and random search are as follows;

[0061]

[0062] Step 5): Merge the new solution and the previous solution and perform non-dominated sorting to obtain the optimal solution at t=1;

[0063] Step 6): Let t+1, use the optimal solution obtained in the tth generation as the parent generation input, repeat steps 2)-4) to generate the offspring;

[0064] Step 7): Merge the parent generation and the child generation, perform fast non-dominated sorting, construct the Pareto set, calculate the congestion, generate a new solution, compare the fitness of the new solution with the current optimal charging station location, and if it is better, replace the current optimal charging station location;

[0065] Step 8): Determine whether the maximum number of iterations has been reached. If not, repeat steps 6) to 7). Otherwise, end the process to obtain the global optimal solution.

[0066] Specifically, after obtaining the above global optimal solution, charging suggestions are sent to users through the transmission and application unit of the terminal to achieve orderly scheduling of regional electric vehicle charging.

[0067] In order to verify the effectiveness of the electric vehicle charging scheduling method based on the improved optimization algorithm proposed in the present invention, data information of a certain area was selected for simulation analysis. The time-of-use electricity price policy of the area is shown in Table 1.

[0068] Table 1 Time-of-use electricity price

[0069]

[0070] Specifically, 178 electric vehicles were selected for simulation analysis, and orderly charging scheduling was carried out for the 178 electric vehicles. The load fluctuations and user charging costs caused by the scheduling were compared with those in the disorderly state. The disorderly charging of electric vehicles is considered to be the start charging time of the electric vehicle's last trip return time of the day. The charging time obeys the normal distribution of 8:00-24:00, and the charging time obeys the normal distribution of 0.5-4 hours. The charging process of the electric vehicle's power battery is approximately considered to be constant power charging.

[0071] Specifically, according to the proposed electric vehicle charging scheduling method based on the improved optimization algorithm, the orderly charging scheme is solved, the orderly charging load curve is obtained, and compared with the disordered charging load curve; in order to verify the effectiveness of the electric vehicle charging scheduling method based on the improved optimization algorithm proposed in the present invention for solving the orderly charging scheme with the traditional optimization algorithm, the NSWOA with elite strategy, the traditional NSWOA, and the NSGA-Ⅱ method proposed in the present invention are used to optimize the scheduling of 178 electric vehicles respectively, and the simulation results are compared to obtain Figure 3 The load curves of orderly charging and disordered charging are shown. It can be seen that all three algorithms show good results.

[0072] Specifically, in the simulation analysis, the parameter values ​​of the load fluctuation degree are set to a=0.4, b=0.3, and c=0.3. The total cost of disordered charging is 1830.90 yuan, and the load fluctuation degree is 492.51. For the solution results of NSWOA with elite strategy, the total cost of user charging is 1536.90 yuan, and the load fluctuation degree is 132.11. For the solution results of traditional NSWOA, the total cost of user charging is 1643.94 yuan, and the load fluctuation degree is 165.69. For the solution results of NSGA-Ⅱ, the total cost of user charging is 1609.44 yuan, and the load fluctuation degree is 172.27. Figure 4Comparison of charging load curves obtained by solving NSWOA with elite strategy and traditional NSWOA. Figure 5 Comparison of charging load curves obtained by solving NSWOA with elite strategy and NSGA-Ⅱ. It can be seen that the electric vehicle charging scheduling method based on NSWOA with elite strategy proposed in the present invention has obvious advantages over other methods.

[0073] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any simple modification, change and equivalent structural change made to the above embodiment based on the technical essence of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. An electric vehicle charging dispatching terminal, characterized in that: In the process of electric vehicle charging scheduling, the changes in the state of charge of the electric vehicle battery are tracked, charging request decisions are made, and corresponding charging suggestions are sent to users. The terminal for electric vehicle charging scheduling includes a data acquisition unit, a data processing and algorithm calculation unit, and a transmission and application unit.

2. An electric vehicle charging scheduling method based on an improved optimization algorithm, characterized in that: The electric vehicle charging dispatching terminal according to claim 1 comprises the following steps: S1. The electric vehicle charging dispatch terminal needs to determine whether the user has issued a charging request based on the charge state of the electric vehicle battery on the user side. If so, proceed to the next step. If not, return to the previous step for loop judgment; S2. The data acquisition unit obtains the daily charging demand and location of electric vehicles, information on charging stations accessible to electric vehicles, regional transportation networks, and operation data of regional distribution networks; S3. Input the acquired relevant data information into the data processing and algorithm calculation unit of the terminal, filter, analyze, extract, model and solve the data through the data processing and algorithm calculation unit to obtain the optimal charging strategy for the electric vehicle; S4. The transmission and application unit of the terminal sends charging plan suggestions to the user to achieve orderly charging scheduling of electric vehicles in the area.

3. According to the electric vehicle charging scheduling method based on the improved optimization algorithm described in claim 2, it is characterized in that: In step S2, the data acquisition unit obtains the daily charging demand and location of the electric vehicle, the information of the charging stations accessible to the electric vehicle, the regional traffic road network, and the operation data of the regional distribution network, which specifically include the charging demand of the electric vehicle within a day, the battery charge state and the location and time information when the electric vehicle issues a charging demand, the location, charging power and efficiency of the charging stations accessible near the area, the time-sharing charging electricity price in the area, the traffic road network map of the area, and the basic load data of the daily operation of the distribution network.

4. According to the electric vehicle charging scheduling method based on the improved optimization algorithm described in claim 2, it is characterized in that: In step S3, the obtained relevant data information is input into the data processing and algorithm calculation unit of the terminal, and the data is filtered, analyzed, extracted, modeled and algorithmically solved by the data processing and algorithm calculation unit to obtain the optimal charging strategy for the electric vehicle, which includes the following steps: S 31 .With the goal of minimizing grid load fluctuation, minimizing the time cost and economic cost of electric vehicle users’ charging, the charging status of electric vehicle users as decision variables, and the charging efficiency of charging stations, the charging capacity of electric vehicles, and the value range of decision variables as constraints, a multi-objective optimization model is established; S 32 .The multi-objective optimization model of electric vehicle charging scheduling is solved using the non-dominated sorting whale optimization algorithm with an elite strategy to obtain the optimal charging strategy.

5. According to claim 4, the electric vehicle charging scheduling method based on the improved optimization algorithm is characterized in that: Step S 31 Specifically include: S 311 The decision variable is defined as x i,j,k , indicating whether the i-th electric car goes to charging station j for charging in the k-th time period. The values ​​of the decision variables are as follows: S 312 .Construct the objective function of multi-objective optimization to minimize the grid load fluctuation F1(x i,j,k ), user charging economic cost F2(x i,j,k ), user charging time cost F3(x i,j,k ), the formula is as follows: Among them, the degree of grid load fluctuation is measured by three indicators: load fluctuation rate, load peak-to-valley difference, and load fluctuation variance. The calculation formulas are as follows: P i，j，k =P max -P min In the formula, B i,j,k represents the load fluctuation rate caused by electric vehicle i charging at charging station j during time period k; P i,j,k represents the peak-to-valley difference of the load when electric vehicle i is charging at charging station j during time period k; P max represents the maximum value of the load fluctuation curve of electric vehicle i after charging at charging station j; P av Represents the average value of the load fluctuation curve; P min Indicates the minimum value of the load fluctuation curve; V i,j,k represents the load variance of electric vehicle i charging at charging station j during period k; The corresponding objective function is: F1(x i，j，k )=x i，j，k *(aB i，j，k +bP i，j，k +cV i，j，k ) In the formula, a, b, c correspond to the weight coefficients of load fluctuation rate, load peak-to-valley difference, and load fluctuation variance, respectively, and a+b+c=1; The economic cost of charging for users is calculated as follows: C1=S j +pr k *t i，j，k In the formula, C1 represents economic cost; S j represents the service fee of charging station j; pr k represents the time-of-use electricity price during period k; t i,j,k represents the charging time of electric vehicle i at charging station j during time period k; The corresponding objective function is F2(x i，j，k )=x i,j,k *C1 The time cost of user charging is considered to be the sum of the time it takes for the electric vehicle to reach the charging station from the departure point, the charging queue time, and the charging time. The calculation formula is as follows: Where C2 represents the time cost of charging, L i,j represents the path length from electric vehicle i to charging station j; v i Indicates the average driving speed; tp j,k represents the queuing time required at charging station j in time period k; ΔC i Indicates the required charging capacity, η j represents the charging efficiency of charging station j; P j represents the charging power provided by charging station j; The corresponding objective function is: F3(x i，j，k )=x i，j,k *C2 S 313 The constraints of the objective function are: In the formula, C i Represents the battery capacity of electric vehicle i.

6. According to claim 4, the electric vehicle charging scheduling method based on the improved optimization algorithm is characterized in that: Step S 32 Specifically include: S 321 . Initialize the location vectors of all N optional charging stations for electric vehicles Calculate the grid load fluctuation, user time cost and economic cost of the currently selected charging station location, and find the current optimal solution F through non-dominated sorting ini (x i,j,k ), and record the location of the optimal charging station at this time as Set the maximum number of iterations to t max , set the current number of iterations t = 1; S 322 . Set the parameter value. First, generate a random number p in [0,1]. If p>0.5, proceed to step S. 323 Otherwise, jump to step S 324 ; Set a=2. During the iteration, a decreases linearly in the range [0,2]. Generate random numbers r1 and r2 in the range [0,1]. Calculate coefficient vectors A and C according to the following formula: A=2ar1-a C＝2r2 S 323 .Perform local search, use the spiral mechanism to update the position, and generate a new position vector. The formula is as follows, where l is a random number in [-1,1]; S 324 . Determine whether |A| is less than 1. If so, update the position through encircling search, otherwise update the position through random search. The formulas for encircling search and random search are as follows: S 325 .Merge the updated solution with the initial solution and perform non-dominated sorting to obtain the optimal solution at t=1; S 326 .Let t+1, use the optimal solution obtained in the tth generation as the parent input, and repeat step S 322 -S 324 , generate offspring; S 327 .Merge the parent generation and the child generation, perform fast non-dominated sorting, construct the Pareto set, calculate the congestion, generate a new solution, compare the fitness of the new solution with the current optimal charging station location, and if it is better, replace the current optimal charging station location; S 328 . Determine whether the maximum number of iterations has been reached. If not, repeat step S 326 -S 327 , otherwise it ends and the global optimal solution is obtained.

Images