Research On Multi-objective Optimal Scheduling Of Electric Vehicle Charging And Discharging In Smart Buildings

In recent years,with the continuous consumption of fossil fuels,the environmental degradation and climate pollution caused by the massive emissions of greenhouse gases have attracted widespread attention around the world.Many countries have begun to formulate corresponding policies to promote the development of new energy industries to reduce the use of fossil fuels.Electric vehicles(EVs)are growing rapidly because they use clean energy.In the future,as the number of EVs grows,the parking lots in or around smart buildings will become the main locations for EVs to park.It is of practical significance to take smart buildings as the main body and unify the dispatch of EVs in or around them through an energy management system,so that EVs can be charged and discharged in an orderly manner,and finally realize the purpose of "peak and valley reduction" in smart buildings’ power grid.Therefore,this paper investigates the multiobjective optimal scheduling of EV charging and discharging in smart buildings from the perspective of smart buildings in terms of EV scheduling model establishment,model solving,algorithm improvement,and model simulation validation.Firstly,this paper describes the development and application prospects of building PV integration technology under the background of "double carbon".On this basis,the composition structure of the smart building microgrid model is introduced,and the concepts of Vehicle to Grid(V2G)and Vehicle to Building(V2B)are explained.Based on the collected travel data of EV owners in Xi’an,the travel characteristics of EVs in different types of smart buildings are analyzed using the Monte Carlo method.Secondly,this paper establishes a charging and discharging scheduling model for EVs in smart buildings.The model is a bi-layer optimization scheduling model,whose upper layer is a multi-objective model with minimum fluctuation of building load and minimum charging cost of EV owners,and the lower layer is a single-objective model with the minimum difference between the total charging capacity of each EV at the end of scheduling and the charging demand of EV owners.For the upper layer model,the battery state constraint of each vehicle as well as the charging and discharging power constraint are considered and solved using the non-dominated ranking genetic algorithmII to obtain the total charging and discharging power of EVs participating in the scheduling under different periods.For the lower layer model,based on the results of the outer layer optimization,the charging and discharging switching times constraint of EVs is considered and solved using the squirrel search algorithm to finally obtain the specific charging and discharging strategy for each EV.Then,this paper proposes a Hybrid random opposition-based learning and Gaussian mutation of the chaotic squirrel search algorithm and solves the lower model according to this algorithm to improve the speed of solving the lower model.The algorithm introduces the Tent chaotic initialization strategy,nonlinear decreasing predator strategy,position greedy selection strategy,Gaussian mutation,and random opposition-based learning in the basic squirrel search algorithm.Through time complexity analysis of the proposed algorithm and performance testing experiments for different types of functions,the results show that the four incorporated improvement strategies not only do not increase the time complexity of the algorithm but also greatly improve the speed and accuracy of the basic squirrel search algorithm for finding the best.The performance of the proposed algorithm has greater superiority in both single-peak and multi-peak test functions.Finally,the historical load data and PV output data of a residential building and an office building are collected for two different types of intelligent buildings in Xi’an,and the data are pre-processed.The simulation results show that the EV charging and discharging scheduling model proposed in this paper reduces the load fluctuation rate in the residential building and the office building by 19.69% and 9.32%,respectively,and the peak-to-valley load difference is reduced by 207.33 k W and 208.33 k W,respectively,in combination with the local time-sharing tariff information in Xi’an;The scheduling model established in this paper regulates EVs in the residential and office buildings in an orderly manner so that EVs discharge when the electricity price is high and charge when the price is low,which not only meets the charging demand of each EV but also reduces the charging cost of EV owners.In summary,the multi-objective optimal scheduling model for EV charging and discharging in smart buildings proposed in this paper can not only realize the purpose of "peak-shaving and valley-filling" in smart building grid,but also reduce the charging cost of EV owners while satisfying their charging demand.It not only alleviates the impact problem when a large number of EVs are connected to the building grid and ensures the safe and stable operation of the building grid but also brings economic benefits to EV owners.