A multi-target-constrained expressway service area new energy vehicle charging path optimization method

Abstract

The application discloses a kind of multi-objective constraint expressway service area new energy vehicle charging path optimization method, comprising: first, the service area micro-grid system model including photovoltaic, energy storage and electric vehicle load is constructed, to minimize the energy load cost of service area as optimization goal, using gauss decreasing inertia weight improved particle swarm algorithm to solve, obtain the optimal energy load cost of service area;Second, based on historical data, vehicle arrival and charging duration distribution are fitted by K-S test, and the queuing theory model is constructed to accurately predict the charging queuing time;Again, the expressway network is modeled by grid method, and the improved A algorithm is combined with directional constraint and mileage limit to plan the shortest path of accessible charging station;Finally, a multi-objective constraint model is established, which takes into account the lowest energy load cost of service area, the shortest queuing time of user, and the shortest driving distance, and uses the improved multi-objective particle swarm optimization algorithm to search for the optimal solution set and recommend the comprehensive optimal charging path.The application can effectively smooth photovoltaic output fluctuation, reduce micro-grid operation cost, reduce charging queuing congestion, improve road energy integration system resource utilization and user charging experience, and adapt to the efficient and low-carbon operation needs of expressway new energy vehicle long-distance supply.

Description

Technical Field

[0001] This invention relates to a method for optimizing charging routes for new energy vehicles, and more particularly to a method for optimizing charging routes for new energy vehicles in highway service areas with multi-objective constraints. Background Technology

[0002] With the increasing number of new energy vehicles, highway service areas have become key charging and replenishment nodes for long-distance travel. Large-scale deployment of charging stations and the promotion of roadside energy integration are crucial for supporting convenient travel for new energy vehicles. To meet the requirements of low-carbon travel, service areas are equipped with clean energy facilities such as photovoltaic systems. However, the current charging scheduling lacks scientific management, resulting in scattered arrival times for transit vehicles and random charging demands, leading to a disorderly connection situation. Coupled with limited charging spaces and unstable clean energy supply, this easily causes problems such as charging queues, congestion, and low efficiency, negatively impacting the user experience.

[0003] In the context of low-carbon transformation of transportation, optimizing charging scheduling and improving service efficiency are the core of perfecting the new energy travel service system. However, the randomness and volatility of vehicle charging demand, combined with the limitations of charging resources and clean electricity supply, result in low resource utilization and long waiting times for users under the traditional model. This not only fails to meet the demand for efficient charging but also hinders the consumption of clean electricity.

[0004] In summary, how to provide a method and device for optimizing charging routes for new energy vehicles in highway service areas under multi-objective constraints has become an urgent problem to be solved in this field. Summary of the Invention

[0005] Purpose of the invention: To address the above problems, this invention proposes a multi-objective constraint-based method for optimizing charging routes for new energy vehicles in highway service areas. The aim is to provide a charging guidance strategy for new energy vehicles on highways to balance grid load, improve facility utilization efficiency, and enhance user experience.

[0006] Technical solution: To achieve the objectives of this invention, the technical solution adopted is as follows: A method for optimizing charging routes for new energy vehicles in highway service areas under multi-objective constraints, comprising the following steps: Step 1: Construct a microgrid system model for the service area and solve it using a Gaussian decreasing inertial weighted particle swarm optimization algorithm to obtain the optimal energy load cost for the service area. Step 2: Based on historical data, identify the distribution of vehicle arrival time intervals and charging duration at service area charging stations, and construct a queuing time prediction model to calculate the queuing time at service area charging stations; Step 3: Model the highway using the grid method and propose an improved directional constraint A. The algorithm obtains the shortest path to the charging station, including: rasterizing the highway network, where each grid cell corresponds to a small area on the map and contains key relational information about that area; and employing an improved A algorithm with Manhattan distance as the core heuristic function. The algorithm incorporates a multi-target parallel search mechanism, remaining drivable mileage constraints, and directional constraints. Step 4: Construct a multi-objective constraint model, search for the optimal solution set based on the improved multi-objective particle swarm optimization algorithm, and recommend the comprehensive optimal charging path. The multi-objective constraint model includes three objective functions: minimizing the energy load cost of the service area, minimizing the user queuing time, and minimizing the driving distance. The improved multi-objective particle swarm optimization algorithm with Gaussian decreasing inertia weights is used to solve for the optimal solution set.

[0007] Specifically, step 1 includes: Microgrids contain loads such as photovoltaics, energy storage devices, and new energy vehicles.

[0008] Photovoltaic power generation is subject to fluctuations and randomness due to factors such as sunlight and weather conditions. The intensity of sunlight throughout the day follows a beta distribution, with the following probability density function: (1) in, It is the intensity of light. and This is a Beta distribution parameter, related to light intensity. and It can be calculated from the average and standard deviation of light intensity over a day: (2) (3) in, and These are light intensity The mean and standard deviation.

[0009] The maximum output power of photovoltaic power is: (4) Where A is the total area of ​​the photovoltaic array. Let be the photoelectric conversion efficiency of the photovoltaic array, and r be the light intensity.

[0010] Based on the characteristics of photovoltaics, a beta distribution is also used to model it: (5) The mathematical model for constructing an energy storage device mainly includes two parameters: battery capacity and state of charge, whose meanings are as follows: (1) Battery capacity The battery capacity of an energy storage device refers to the amount of electrical energy that can be completely released through a discharge operation when the energy storage device is fully charged. The amount of electrical energy released at this time is the battery capacity, also known as the rated capacity.

[0011] (2) State of charge (SOC)

[0012] State of Charge (SOC) refers to the ratio of the remaining electrical capacity of an energy storage device to its rated capacity at a given moment, typically ranging from 0 to 1. Energy storage devices generally operate in two states: charging and discharging. The SOC model for these two states is as follows: (6) In the formula, This indicates the SOC state at this moment. This indicates the SOC state at the previous moment. Indicates the charging and discharging power of the energy storage device. , This indicates the charging and discharging efficiency of energy storage devices.

[0013] The charging and discharging power and state of charge of energy storage devices must be within a certain upper and lower limit: (7) (8) The probability density function of the charging start time of new energy vehicles follows a normal distribution: (9) in, =15, =3.4.

[0014] The probability density function of the daily mileage of a vehicle follows a log-normal distribution: (10) in, =3.2, =0.88.

[0015] A disordered charging model for new energy vehicles was established, and the charging time can be obtained using the following formula: (11) Where S represents the daily mileage. Electricity consumption per 100 kilometers Charging power for new energy vehicles, Improve the charging efficiency of new energy vehicles.

[0016] For the economic dispatch optimization of microgrids in highway service areas, the minimum energy load cost is selected as the objective function, including the cost of energy storage equipment, photovoltaic power generation subsidies, and the electricity purchase and sale costs between the microgrid and the external main grid. The specific objective function is... The expression is as follows: (12) In the formula, F represents the economic cost. For electricity purchase costs, For electricity sales costs, For the cost of energy storage equipment, For photovoltaic operation and maintenance costs, Penalty costs for grid interaction.

[0017] The constraints are as follows: (1) Power balance constraint This energy system must ensure that the sum of the output power from the external power grid, energy storage devices, and photovoltaic power generation meets the basic electricity load and new energy vehicle load demands of the service area, which can be expressed as: (13) In the formula, For the service area's basic load, For the load of new energy vehicles, This refers to the electrical power purchased from the grid during grid interconnection. For the power of energy storage devices, This refers to the photovoltaic power generation capacity.

[0018] (2) Power constraints of grid interaction (14) In the formula, This is the upper limit of the power exchange between the service area microgrid and the external power grid.

[0019] (3) Constraints of energy storage equipment

[0020] ①SOC constraint. The state of charge of the energy storage device is between the upper and lower limits at any given time.

[0021] (15)

[0022] In the formula, The state of charge of an energy storage device at a certain moment.

[0023] ② Energy storage energy balance constraints. The initial and final SOC remain consistent, facilitating the recycling of energy storage.

[0024] (16)

[0025] In the formula, Let t be the state of charge of the energy storage device; This represents the initial state of charge of the energy storage device.

[0026] (3) Photovoltaic constraint (17) In the formula, Photovoltaic power generation capacity, , These represent the minimum and maximum photovoltaic power generation values, respectively.

[0027] An improved particle swarm optimization algorithm with Gaussian decreasing inertia weights is used to solve the problem. The inertia weights are improved by using non-linearly decreasing Gaussian weights, and are calculated as follows: (18) in, As the inertia weighting factor, This represents the current iteration number. This represents the maximum number of iterations.

[0028] The learning factor is also improved and calculated as follows: (19) (20) in, This is the current iteration number; It represents the maximum number of iterations.

[0029] The main steps of the improved particle swarm optimization algorithm with Gaussian decreasing inertia weights are as follows: (1) Initialize the particle swarm. Determine the population size and the initial position of each particle. and speed .

[0030] (2) Calculate the fitness of the particles. In this study, fitness is defined as economic cost. .

[0031] (3) Obtain and In particle swarm optimization algorithms, This refers to the historical optimal position information of each particle obtained through the individual's optimal position, while This refers to selecting a group-optimal position from the historical optimal positions of these individuals and using it as the current historical optimal position.

[0032] (4) Update the state. Velocity vector The update is as follows: (twenty one) Position vector The update is as follows: (twenty two) in .

[0033] (5) Boundary condition handling. Update the position vector. When doing so, it is necessary to restrict the position of the particles to the boundary of the search space.

[0034] (6) Determine the termination condition. Check the iteration count: Determine the current iteration count. Has the maximum number of iterations been reached? If the current number of iterations has reached the maximum number of iterations, terminate the iteration; check for convergence: check if the fitness value meets the convergence condition. If the convergence condition is met, the algorithm terminates and outputs the global optimal solution.

[0035] (7) Output results. Obtain the optimal energy load cost of the microgrid in the service area.

[0036] Specifically, step 2 includes: The distribution of vehicle arrival time intervals and charging duration at service area charging stations was obtained by fitting actual data through the KS test. Based on the distribution of vehicle arrival time intervals and charging durations, parameter values ​​for each time period are obtained using actual data to simulate the arrival of new energy vehicles at charging stations during different time periods. The process of adding and removing vehicles within the charging station queuing system includes two scenarios: new energy vehicles arriving and entering the queue, and new energy vehicles completing charging and leaving the charging station. A queuing model based on actual conditions is constructed based on these two events. (1) The next event is the arrival of the vehicle. Update the current time when a new energy vehicle arrives at a charging station. This refers to the arrival time.

[0037] If the charging station has no available charging guns at this time, the new energy vehicle joins the queue and its arrival time is recorded. If there is an available charging gun at the charging station, the new energy vehicle will begin charging directly, and the charging time will be randomly obtained from a normal distribution of charging times. Update charging completion time : (twenty three) The queuing time for new energy vehicles at this time It is 0.

[0038] (2) The next event is the completion of charging.

[0039] Update current time : (twenty four) If no vehicle is waiting at this time, the charging gun is marked as idle.

[0040] If there are vehicles waiting, the vehicle at the front of the queue will begin charging, and the charging completion time will be updated simultaneously. This is the queue time for new energy vehicles. for: (25) Specifically, step 4 includes: The multi-objective constraint model includes three objective functions: minimizing the energy load cost of the service area, minimizing the user queuing time, and minimizing the driving distance. The difference in optimal energy load cost before and after new energy vehicles are added to the service area microgrid is taken as the first objective function, and the formula is as follows: (26) In the formula, This represents objective function one, namely the energy load cost difference of the microgrid in the service area; i Indicates the candidate highway service area number; This indicates that when the new energy vehicle is not connected, the first i The optimal energy load cost for each service area during the current time period; This indicates that after the new energy vehicle is connected, the first i The optimal energy load cost is obtained by re-optimizing the scheduling of each service area.

[0041] Queuing time objective function Defined as the queuing time of the vehicle in the candidate service area: (27) In the formula, This indicates that the vehicle v obtained by using the queuing model logic described in step 2 is... Time reaches the i The actual waiting time after each service area.

[0042] Objective function of driving distance Defined as the length of the optimal path: (28) In the formula, Indicates improvement A The shortest path length output by the algorithm.

[0043] An improved multi-objective particle swarm optimization algorithm with Gaussian decreasing inertia weights is used to solve for the optimal solution set. The algorithm process consists of the following steps: (1) Complete the configuration of basic algorithm parameters, including population size, maximum number of iterations and storage capacity of external archive; (2) Calculate the fitness value of the initial population, obtain the non-dominated solutions and add them to the external archive set; (3) Update the position and velocity of the particles; (4) Calculate the particle's current fitness value and... Compare, maintain and update external archives. fitness value; (5) Determine whether the termination condition is met. If not, continue the loop. If it is met, output the external archive set.

[0044] The working principle of this invention is as follows: This invention integrates four core components: microgrid economic dispatch, charging queuing prediction, path planning, and multi-objective optimization. First, a service area microgrid model incorporating photovoltaic, energy storage, and new energy vehicle loads is constructed. With the goal of minimizing the energy load cost of the service area, a Gaussian decreasing inertia weighted particle swarm optimization algorithm is used to solve the problem, yielding the optimal energy load cost for the service area. Second, historical charging data is used to fit the distribution of vehicle arrival intervals and charging time through the KS test, constructing an event-driven queuing model to calculate the charging queuing time at each service area charging station. Finally, a grid method is used to perform gridded modeling of the highway network, building upon traditional A / B algorithm... Based on the algorithm, directional constraints, remaining mileage limits, and a multi-objective point parallel search mechanism are incorporated to achieve the shortest driving path planning to reach the charging station. Finally, a multi-objective constraint model is constructed to minimize the energy load cost of the service area, the user queuing time, and the driving distance. The Pareto optimal solution set is solved by improving the multi-objective particle swarm algorithm, and the comprehensive optimal charging path is recommended.

[0045] The innovation of this invention is: 1) Focusing on the specific energy replenishment scenario of highway service areas, a collaborative scheduling architecture covering multiple dimensions has been established.

[0046] 2) The nonlinear modeling of particle swarm inertia weights using a Gaussian decreasing function provides a more accurate solution for multi-objective optimization of microgrids.

[0047] 3) The queuing model was built based on actual queuing data, which significantly enhanced the model's ability to solve the problems of real traffic flow and charging load fluctuations.

[0048] The advantages of this invention are: 1) This invention delves into the energy replenishment characteristics of highway service areas in a specific scenario, deeply coupling the one-way topology of highways, the hard constraint of vehicle remaining range (SOC), and the "source-load" fluctuations of the service area microgrid. By establishing a multi-dimensional constraint framework, it solves the problem of mismatch between spatial accessibility and energy management in long-distance energy replenishment scenarios.

[0049] 2) This invention introduces a Gaussian decreasing function to nonlinearly model the inertia weights of the particle swarm optimization algorithm, effectively balancing the algorithm's global exploration capability in the early stages of the search with its local search accuracy in the later stages. Under complex constraints, it exhibits stronger optimization stability and convergence speed, providing higher-precision computational support for microgrid economic analysis and multi-objective optimization.

[0050] 3) This invention overcomes the limitation of traditional queuing theory research that assumes vehicle arrival and charging times follow an ideal distribution. By statistically fitting actual operational data from highway service areas, a queuing model based on real-world queuing conditions is constructed. It exhibits extremely high simulation fit in predicting charging waiting times, accurately extrapolating dynamic queuing pressure under extreme peak traffic conditions, and significantly improving the effectiveness and accuracy of scheduling decisions. Attached Figure Description

[0051] Figure 1 This is a flowchart of the process of this invention; Figure 2 This is a Gaussian decreasing inertia weight map of the improved particle swarm optimization algorithm in an embodiment of the present invention; Figure 3 This is a service area queuing model diagram according to an embodiment of the present invention.

[0052] Figure 4 This is a grid road network diagram according to an embodiment of the present invention.

[0053] Figure 5 This is a flowchart of the multi-objective particle swarm algorithm in an embodiment of the present invention.

[0054] Figure 6 This is a multi-objective constrained optimal charging path diagram in an embodiment of the present invention. Detailed Implementation

[0055] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0056] A multi-objective constrained method for optimizing charging routes for new energy vehicles in highway service areas includes: Step 1: Construct a microgrid system model for the service area and solve it using a Gaussian decreasing inertial weighted particle swarm optimization algorithm to obtain the optimal energy load cost for the service area. Step 2: Based on historical data, identify the distribution of vehicle arrival time intervals and charging duration at service area charging stations, and construct a queuing time prediction model to calculate the queuing time at service area charging stations; Step 3: Model the highway using the grid method and propose an improved directional constraint A. The algorithm obtains the shortest path to the charging station, including: rasterizing the highway network, where each grid cell corresponds to a small area on the map and contains key relational information about that area; and employing an improved A algorithm with Manhattan distance as the core heuristic function. The algorithm incorporates a multi-target parallel search mechanism, remaining drivable mileage constraints, and directional constraints. Step 4: Construct a multi-objective constraint model, search for the optimal solution set based on the improved multi-objective particle swarm optimization algorithm, and recommend the comprehensive optimal charging path. The multi-objective constraint model includes three objective functions: minimizing the energy load cost of the service area, minimizing the user queuing time, and minimizing the driving distance. The improved multi-objective particle swarm optimization algorithm with Gaussian decreasing inertia weights is used to solve for the optimal solution set.

[0057] Specifically, step 1 includes: Microgrids contain loads such as photovoltaics, energy storage devices, and new energy vehicles.

[0058] Photovoltaic power generation is subject to fluctuations and randomness due to factors such as sunlight and weather conditions. The intensity of sunlight throughout the day follows a beta distribution, with the following probability density function: (1) in, It is the intensity of light. and This is a Beta distribution parameter, related to light intensity. and It can be calculated from the average and standard deviation of light intensity over a day: (2) (3) in, and These are light intensity The mean and standard deviation.

[0059] The maximum output power of photovoltaic power is: (4) Where A is the total area of ​​the photovoltaic array. Let be the photoelectric conversion efficiency of the photovoltaic array, and r be the light intensity.

[0060] Based on the characteristics of photovoltaics, a beta distribution is also used to model it: (5) The mathematical model for constructing an energy storage device mainly includes two parameters: battery capacity and state of charge, whose meanings are as follows: (1) Battery capacity The battery capacity of an energy storage device refers to the amount of electrical energy that can be completely released through a discharge operation when the energy storage device is fully charged. The amount of electrical energy released at this time is the battery capacity, also known as the rated capacity.

[0061] (2) State of charge (SOC)

[0062] State of Charge (SOC) refers to the ratio of the remaining electrical capacity of an energy storage device to its rated capacity at a given moment, typically ranging from 0 to 1. Energy storage devices generally operate in two states: charging and discharging. The SOC model for these two states is as follows: (6) In the formula, This indicates the SOC state at this moment. This indicates the SOC state at the previous moment. Indicates the charging and discharging power of the energy storage device. , This indicates the charging and discharging efficiency of energy storage devices.

[0063] The charging and discharging power and state of charge of energy storage devices must be within a certain upper and lower limit: (7) (8) The probability density function of the charging start time of new energy vehicles follows a normal distribution: (9) in, =15, =3.4.

[0064] The probability density function of the daily mileage of a vehicle follows a log-normal distribution: (10) in, =3.2, =0.88.

[0065] A disordered charging model for new energy vehicles was established, and the charging time can be obtained using the following formula: (11) Where S represents the daily mileage. Electricity consumption per 100 kilometers Charging power for new energy vehicles, Improve the charging efficiency of new energy vehicles.

[0066] For the economic dispatch optimization of microgrids in highway service areas, the minimum energy load cost is selected as the objective function, including the cost of energy storage equipment, photovoltaic power generation subsidies, and the electricity purchase and sale costs between the microgrid and the external main grid. The specific objective function is... The expression is as follows: (12) In the formula, F represents the economic cost. For electricity purchase costs, For electricity sales costs, For the cost of energy storage equipment, For photovoltaic operation and maintenance costs, Penalty costs for grid interaction.

[0067] The constraints are as follows: (1) Power balance constraint This energy system must ensure that the sum of the output power from the external power grid, energy storage devices, and photovoltaic power generation meets the basic electricity load and new energy vehicle load demands of the service area, which can be expressed as: (13) In the formula, For the service area's basic load, For the load of new energy vehicles, This refers to the electrical power purchased from the grid during grid interconnection. For the power of energy storage devices, This refers to the photovoltaic power generation capacity.

[0068] (4) Power constraints of grid interaction (14) In the formula, This is the upper limit of the power exchange between the service area microgrid and the external power grid.

[0069] (3) Constraints of energy storage equipment

[0070] ①SOC constraint. The state of charge of the energy storage device is between the upper and lower limits at any given time.

[0071] (15)

[0072] In the formula, The state of charge of an energy storage device at a certain moment.

[0073] ② Energy storage energy balance constraints. The initial and final SOC remain consistent, facilitating the recycling of energy storage.

[0074] (16)

[0075] In the formula, Let t be the state of charge of the energy storage device; This represents the initial state of charge of the energy storage device.

[0076] (5) Photovoltaic confinement (17) In the formula, Photovoltaic power generation capacity, , These represent the minimum and maximum photovoltaic power generation values, respectively.

[0077] An improved particle swarm optimization algorithm with Gaussian decreasing inertia weights is used to solve the problem. The inertia weights are improved by using non-linearly decreasing Gaussian weights, such as... Figure 2 As shown, the calculation is as follows: (18) in, As the inertia weighting factor, This represents the current iteration number. This represents the maximum number of iterations.

[0078] The learning factor is also improved and calculated as follows: (19) (20) in, This is the current iteration number; It represents the maximum number of iterations.

[0079] The main steps of the improved particle swarm optimization algorithm with Gaussian decreasing inertia weights are as follows: (1) Initialize the particle swarm. Determine the population size and the initial position of each particle. and speed .

[0080] (2) Calculate the fitness of the particles. In this study, fitness is defined as economic cost. .

[0081] (3) Obtain and In particle swarm optimization algorithms, This refers to the historical optimal position information of each particle obtained through the individual's optimal position, while This refers to selecting a group-optimal position from the historical optimal positions of these individuals and using it as the current historical optimal position.

[0082] (4) Update the state. Velocity vector The update is as follows: (twenty one) Position vector The update is as follows: (twenty two) in .

[0083] (5) Boundary condition handling. Update the position vector. When doing so, it is necessary to restrict the position of the particles to the boundary of the search space.

[0084] (6) Determine the termination condition. Check the iteration count: Determine the current iteration count. Has the maximum number of iterations been reached? If the current number of iterations has reached the maximum number of iterations, terminate the iteration; check for convergence: check if the fitness value meets the convergence condition. If the convergence condition is met, the algorithm terminates and outputs the global optimal solution.

[0085] (7) Output results. Obtain the optimal energy load cost of the microgrid in the service area.

[0086] This embodiment selects the Xiaoshan Expressway service area in Hangzhou City, Zhejiang Province.

[0087] The data in this embodiment includes parameters related to photovoltaic equipment, parameters related to energy storage equipment, parameters related to the main grid, and time-of-use electricity prices, as shown in Tables 1, 2, 4, and 3.

[0088] Table 1

[0089] Table 2

[0090] Table 3

[0091] Table 4

[0092] Specifically, step 2 includes: The distribution of vehicle arrival time intervals and charging duration at service area charging stations was obtained by fitting actual data through the KS test. Based on the distribution of vehicle arrival time intervals and charging durations, parameter values ​​for each time period are obtained using actual data to simulate the arrival of new energy vehicles at charging stations during different time periods. The process of adding and removing vehicles within the charging station queuing system includes two scenarios: new energy vehicles arriving and entering the queue, and new energy vehicles completing charging and leaving the charging station. A queuing model based on these two events is constructed, such as... Figure 3 As shown; (1) The next event is the arrival of the vehicle. Update the current time when a new energy vehicle arrives at a charging station. This refers to the arrival time.

[0093] If the charging station has no available charging guns at this time, the new energy vehicle joins the queue and its arrival time is recorded. If there is an available charging gun at the charging station, the new energy vehicle will begin charging directly, and the charging time will be randomly obtained from a normal distribution of charging times. Update charging completion time : (twenty three) The queuing time for new energy vehicles at this time It is 0.

[0094] (2) The next event is the completion of charging.

[0095] Update current time : (twenty four) If no vehicle is waiting at this time, the charging gun is marked as idle.

[0096] If there are vehicles waiting, the vehicle at the front of the queue will begin charging, and the charging completion time will be updated simultaneously. This is the queue time for new energy vehicles. for: (25) The distribution of vehicle arrival time intervals and charging time in this embodiment is shown in Table 5.

[0097] Table 5. Distribution of arrival time interval and charging duration of new energy vehicles

[0098] Specifically, step 3 includes: For the rasterization of the highway network, each grid cell corresponds to a small area on the map and contains key related information about that area, such as... Figure 4 As shown; An improved A algorithm using Manhattan distance as the core heuristic function is adopted. The algorithm incorporates a multi-target parallel search mechanism, remaining drivable mileage constraints, and directional constraints.

[0099] Specifically, step 4 includes: The multi-objective constraint model includes three objective functions: minimizing the energy load cost of the service area, minimizing the user queuing time, and minimizing the driving distance. The difference in optimal energy load cost before and after new energy vehicles are added to the service area microgrid is taken as the first objective function, and the formula is as follows: (26) In the formula, This represents objective function one, namely the energy load cost difference of the microgrid in the service area; i Indicates the candidate highway service area number; This indicates that when the new energy vehicle is not connected, the first i The total cost of optimal microgrid energy load for each service area during the current time period; This indicates that after the new energy vehicle is connected, the first i The optimal total cost of microgrid energy load is obtained by re-optimizing the scheduling of each service area.

[0100] Queuing time objective function Defined as the queuing time of the vehicle in the candidate service area: (27) In the formula, This indicates that the vehicle v obtained by using the queuing model logic described in step 2 is... Time reaches the i The actual waiting time after each service area.

[0101] Objective function of driving distance Defined as the length of the optimal path: (28) In the formula, Indicates improvement A The shortest path length output by the algorithm.

[0102] An improved multi-objective particle swarm optimization algorithm using Gaussian decreasing inertia weights is used to solve for the optimal solution set, such as... Figure 5 As shown, the algorithm flow consists of the following steps: (1) Complete the configuration of basic algorithm parameters, including population size, maximum number of iterations and storage capacity of external archive; (2) Calculate the fitness value of the initial population, obtain the non-dominated solutions and add them to the external archive set; (3) Update the position and velocity of the particles; (4) Calculate the particle's current fitness value and... Compare, maintain and update external archive sets. fitness value; (5) Determine whether the termination condition is met. If not, continue the loop. If it is met, output the external archive set.

[0103] Taking the highway network at the junction of Hangzhou, Jiaxing, and Shaoxing in Zhejiang Province and surrounding areas as an example, the optimal charging route obtained by a certain new energy vehicle is as follows: Figure 6 As shown.

[0104] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention. The functional modules described above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any way. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent functional substitutions, and improvements made to the above embodiments within the spirit and principles of the present invention, without departing from the content of the present invention, are still within the protection scope of the present invention.

Claims

1. A method for optimizing charging routes for new energy vehicles in highway service areas under multi-objective constraints, characterized in that, Includes the following steps: Step 1: Construct a microgrid system model for the service area and solve it using a Gaussian decreasing inertial weighted particle swarm optimization algorithm to obtain the optimal energy load cost for the service area. Step 2: Based on historical data, identify the distribution of vehicle arrival time intervals and charging duration at service area charging stations, and construct a queuing time prediction model to calculate the queuing time at service area charging stations; Step 3: Model the highway using the grid method and propose an improved directional constraint A. The algorithm obtains the shortest path to the charging station, including: rasterizing the highway network, where each grid cell corresponds to a small area on the map and contains key relational information about that area; and employing an improved A algorithm with Manhattan distance as the core heuristic function. The algorithm incorporates a multi-target parallel search mechanism, remaining drivable mileage constraints, and directional constraints. Step 4: Construct a multi-objective constraint model, search for the optimal solution set based on the improved multi-objective particle swarm optimization algorithm, and recommend the comprehensive optimal charging path. The multi-objective constraint model includes three objective functions: minimizing the energy load cost of the service area, minimizing the user queuing time, and minimizing the driving distance. The improved multi-objective particle swarm optimization algorithm with Gaussian decreasing inertia weights is used to solve for the optimal solution set.

2. The method for optimizing charging routes for new energy vehicles in highway service areas under multi-objective constraints according to claim 1, characterized in that, In step 1, a service area microgrid system model is constructed; The microgrid includes loads such as photovoltaic (PV) power generation, energy storage devices, and new energy vehicles. A PV power generation model, an energy storage model, and a model for disordered charging of new energy vehicles are established. The objective function is to minimize the energy load cost of the service area, including the cost of energy storage equipment, photovoltaic power generation subsidies, and the cost of purchasing and selling electricity between the microgrid and the external main grid. Constraints include power balance constraints, grid interaction power constraints, energy storage equipment constraints, and photovoltaic constraints.

3. The method for optimizing charging routes for new energy vehicles in highway service areas under multi-objective constraints according to claim 1, characterized in that, In step 1, the particle swarm optimization algorithm with Gaussian decreasing inertia weights is used to solve the problem. The particle swarm optimization algorithm is improved by adjusting the inertia weight and learning factor. The inertia weight uses a non-linearly decreasing Gaussian decreasing method, calculated as follows: in, As the inertia weighting factor, This represents the current iteration number. This represents the maximum number of iterations.

4. The method for optimizing charging routes for new energy vehicles in highway service areas under multi-objective constraints according to claim 1, characterized in that, In step 2, the distribution of vehicle arrival time intervals and charging durations at service area charging stations is determined based on historical data: The distribution of vehicle arrival time intervals and charging duration at service area charging stations was obtained by fitting actual data through the KS test.

5. The method for optimizing charging routes for new energy vehicles in highway service areas under multi-objective constraints according to claim 1, characterized in that, In step 2, a queuing time prediction model is constructed: Based on the distribution of vehicle arrival time intervals and charging durations, parameter values ​​for each time period are obtained using actual data to simulate the arrival of new energy vehicles at charging stations at different times. The process of adding and removing vehicles in a charging station queuing system involves two scenarios: new energy vehicles arriving and entering the queue, and new energy vehicles completing charging and leaving the charging station. Based on these two events, a queuing model based on actual conditions is constructed to calculate the queuing waiting time for new energy vehicles within the station.

6. The method for optimizing charging routes for new energy vehicles in highway service areas under multi-objective constraints according to claim 1, characterized in that, In step 4, a multi-objective constraint model is constructed, and an improved multi-objective particle swarm optimization algorithm with Gaussian decreasing inertia weights is used to solve for the optimal solution set. The algorithm process consists of the following steps: (1) Complete the configuration of basic algorithm parameters, including population size, maximum number of iterations and storage capacity of external archive; (2) Calculate the fitness value of the initial population, obtain the non-dominated solutions and add them to the external archive set; (3) Update the position and velocity of the particles; (4) Calculate the particle's current fitness value and... Compare, maintain and update external archive sets. fitness value; (5) Determine whether the termination condition is met. If not, continue the loop. If it is met, output the external archive set.

Images