Research On Swarm Intelligence Optimization Algorithm For Bidirectional Charge-discharge Of Microgrid Model With Electric Vehicles

As a powerful supplement to the power grid system,the microgrid(MG)represents the development trend of future energy,and is an important carrier to promote the efficient complementary use of renewable energy and promote the sustainable development strategy.At the same time,with the continuous improvement of the proportion of electric vehicles in the automobile market,electric vehicles are integrated into the MG.Because of the randomness of its charging and discharging rules,it will have a significant impact on the operation of the MG and pose a huge burden on the power system,resulting in the expansion of the peak-valley difference of the MG load distribution and the increase of economic costs.At present,the charging and discharging period of electric vehicles is mainly controlled by the time-of-use electricity price policy.However,due to the mismatch between the original load curve of the MG and the demand of the time-of-use electricity price interval,a large number of electric vehicles may be charged during the peak load period,resulting in excessive load pressure on the MG.Therefore,this paper aims to design a good performance swarm intelligence optimization algorithm from the perspective of optimization to apply to the current MG scheduling optimization problem under vehicle-to-grid(V2G).Based on the widely used MG scheduling optimization model,this paper analyzes the advantages and disadvantages of some existing swarm intelligence optimization algorithms through test function sets,and proposes an improved sparrow search algorithm(ISSA)based on the characteristics of the MG scheduling optimization model.The algorithm is successfully applied to the MG scheduling optimization problem with different Charging and discharging control strategy for electric vehicles.The main research work is as follows:(1)Considering a grid-connected MG including wind turbine,photovoltaic,microturbine,fuel cell,electric vehicle and batteries,the mechanism model of interactive power of these distributed devices is analyzed in turn,and how to schedule these devices to achieve the expected purpose of MG scheduling optimization is introduced.The MG scheduling is transformed into a mathematical optimization model,and the corresponding decision variables,objective functions and constraints of the optimization problem are given.This paper lists two main ways for electric vehicles to enter the network.By showing the shortcomings of these methods,it is proposed to add the charging and discharging time period of electric vehicles to the decision variables that need to be scheduled in the MG.The optimization algorithm is used to realize the bi-level optimization.On the basis of optimizing the charging and discharging time period of electric vehicles,the production plan of each distributed device in the MG is optimized to adapt to the mismatch between the time-of-use price period and the load curve.(2)In order to apply to high-dimensional optimization problems such as MG scheduling optimization problems,this paper first proposes an ISSA based on particle swarm optimization algorithm with hierarchical speed.The core idea is to archive the historical trajectory of sparrow individuals as producers,so that these sparrow individuals learn the local exploration ability of particle swarm optimization algorithm.In addition,the convergence speed of the improved algorithm is improved by introducing the group better producer following strategy into the follower updating formula.In the algorithm performance comparison experiment,the ISSA proposed in this paper is compared with the sparrow search algorithm(SSA),whale optimization algorithm(WOA),particle swarm optimization algorithm(PSO)and gravitational search algorithm(GSA).Fifteen benchmark functions were selected in the CEC-2017 benchmark functions,and were divided into three groups according to the characteristics of the benchmark function.These test functions were optimized to verify the performance of the optimization algorithm.The mean,optimal value,standard deviation and convergence curve of the optimized test functions are regarded as evaluation indexes to verify the performance of the optimization algorithm.(3)The application of five swarm intelligent optimization algorithms including ISSA to the MG scheduling optimization under different EV charging and discharging control scenarios is studied,and the influence of these scenarios on the economic operating costs obtained from the MG scheduling optimization is analyzed.Experimental results show that compared with other swarm intelligent optimization algorithms,the ISSA proposed in this paper can achieve better performance in MG scheduling optimization problems,and can adapt to different EV charging and discharging control scenarios.The charging and discharging time period optimization strategy of electric vehicles proposed in this paper can adapt to the high economic operation cost of micro-grid caused by the mismatch between load curve and the TOU electricity price interval on the basis of the TOU electricity time period,and provide a more reasonable reference interval for the charging and discharging time period management of electric vehicles.