Coordinated Optimization Between Battery Swapping Station And Microgrid With Distributed Generation

With the depletion of fossil energy and global warming and other environmental problems becoming increasingly serious,the development and utilization of renewable energy such as wind and solar energy is of great significance for changing China's energy structure,ensuring energy security and improving energy efficiency.Micro-grid integrates Distributed Generation(DG),energy storage and load,such as wind power and photovoltaic power,into a whole.With Electric Vehicle(EV)becoming more and more popular,the Battery Swapping Station(BSS)is spawning.More and more BSS have been put into operation in microgrid to meet the demand of EV quick charging and switching.When the Micro-grid containing Distributed Generation(DG)is operating,due to the randomness and uncertainty of DG output and the addition of battery changing station,the operation of the Micro-grid becomes more complex and unpredictable.It is of great practical significance and engineering value to coordinate and optimize the operation between BSS and the micro-grid containing DG,so as to improve the utilization rate of the micro-grid to DG,realize the mutual benefit and win-win of the two independent subjects,BSS and the micro-grid containing DG,and achieve the ultimate goal of reducing environmental pollution.This paper mainly includes two aspects: the research on renewable energy output and power load prediction,and the construction of the coordination and optimization mathematical model between the distributed generation and BSS and the research on its solution method.According to the characteristics of wind,optical power generation and power load,NARX neural network was used to predict the output and power load of renewable energy,because the wind,optical output and power load are highly nonlinear,time-varying and uncertain.Put forward a forecasting method based on improved Harmony Search(HS)algorithm to optimize NARX neural network,improve the conventional NARX neural network easy to fall into local optimum,use the improved HS algorithm to optimize NARX neural network weights and threshold values,accelerate the NARX neural network convergence speed and avoid the premature convergence problem,improve the accuracy of prediction,improve utilization of renewable energy in the micro grid,reduce the cost of the micro grid operation.According to the BSS and micro Power grid that containing DG coordination operation optimization,the paper from three aspects to study:(1)Put forward a kind of micro Power grid containing DG bi-level optimization model of operation in coordination with BSS,this model considers the and micro grid topology structure of BSS integration,build the micro grid economic operation of AC Optimal Power Flow optimization(AC-OPF)model and BSS economic dispatching of a Mixed Integer Linear Programming(MILP)model.(2)If the two models integrate directly,a complex mixed integer nonlinear problem with long computation time will be generated.Therefore,two methods,peer-to-peer(P2P)and leader-follower(LF),are proposed: for P2 P method,the internal parameters of microgrid and BSS are confidential,and only a single shadow price and power transaction request need to be exchanged between the two entities.This method is applicable to the cooperative operation of microgrid and BSS as two independent operators.With the LF approach,the microgrid entity has access to the internal data of BSS,while the BSS entity can make independent decisions on energy management.This method is applicable to the microgrid and BSS are not independent operators,the microgrid can access all the data inside BSS.These two methods can be adapted to a variety of scenarios to achieve coordinated optimization of BSS and microgrid operation.(3)P2P and LF methods are further improved.For the P2 P method,only the shadow price of the individual and the charging and discharging power between the two entities are exchanged in the iterative process.For LF method,the improved column-and-con-straint Generation(C&CG)method is used to solve the bilayer optimization problem,and the original problem is decomposed into a main problem and a sub-problem for iterative solution.Finally,through the example analysis of IEEE 33 bus system,the numerical study shows that the prediction method of NARX neural network optimized based on the improved HS algorithm can predict the wind and light output more accurately and quickly,improve the utilization rate of renewable energy,and reduce the operation cost of microgrid with distributed generation.The effectiveness of P2 P and LF methods is verified by the example analysis of IEEE 33 bus system integrated with BSS,and the advantages of the proposed method in reducing the total operation cost are verified by comparing with the direct solution method.Finally,the fast convergence of the two methods is proved by the finite number of iterations and the computation time.