Research On Multi-layer Optimization Strategy Of Dynamic Ride-sharing Based On Charging Pile Sharing

In order to cope with traffic congestion,energy consumption,environmental pollution and other issues,local governments actively promote new energy vehicles and vigorously build basic charging facilities;major travel platforms actively explore new modes of transportation and promote the sharing of private charging piles.In this context,the new energy electric vehicle sharing came into being.It quickly matched the supply and demand request,increased the average passenger capacity,increased the travel platform revenue,reduced passenger travel costs,effectively alleviated traffic congestion,reduced environmental pollution,and created great social value and welfare.The new energy electric vehicle sharing mode mainly faces two problems of vehicle path planning and electric vehicle charging.Effective vehicle path planning can reduce the cost of the shared carrier.A reasonable vehicle charging scheduling strategy can reduce the standard deviation of the charging pile grid load.Reducing the maintenance cost of charging pile operators,how to design a complete vehicle scheduling solution to balance the goals of the shared carrier and charging station operators has become the focus of current research.Based on the commuter sharing and distributed shared charging pile model in the city,this paper introduces the participants of the shared carrier,charging station operator,electric vehicle manufacturer and government regulatory department,and proposes a dynamic sharing mode based on charging pile sharing.Supporting system function modules,operating mechanisms,profit models,reward models,etc.,to ensure the feasibility and sustainability of the model.Taking Wuhan as an example,this paper predicts the car ownership through principal component analysis and other methods,introduces the new product diffusion model of Bass,studies the diffusion ratio of electric vehicles in automobiles,and then obtains the forecast data of electric vehicle ownership,and obtains the vehicle by combining the charging pile data.The pile ratio predicts the results to verify the validity of the model.Based on the dynamic sharing model based on charging pile sharing,this paper designs a two-layer optimization model.The upper layer studies the dynamic sharing operation cost minimization.The lower layer studies the charging standard deviation of the grid.The upper and lower layers interact with each other.The model seeks a Pareto optimal solution set between the two objective functions.In order to solve the two-layer optimization model,this paper designs an improved NSGA-II algorithm,introduces the vehicle path identifier in the chromosome coding,initializes the population by the point insertion method,and replaces the crossover operator with the same method to avoid the excellent individual iteration in the population.After becoming an invalid solution,the acceleration algorithm converges.In this paper,based on the two subsets of r101 and c101 of Solomon dataset,the simulation experiment is carried out.The experimental results show that the proposed two-layer optimization model and improved NSGA-II algorithm have algorithm performance,vehicle-pile ratio,optimized path,and shared operation cost.The standard deviation of the charging pile load is good,and it is suitable for dealing with large-scale demand scenarios.