Research On Energy Optimized Operation Of Multi-Microgrid Based On Game Theort

As a novel distribution network structure,the microgrid(MG)can fully utilize the application potential of distributed energy and has become an important part of the smart grid in the future.The growth of distributed power sources has also caused more and more MGs to appear in distribution networks.The trading mode and coordinated control of multi-MG are the key to affecting MG groups and regional distribution networks.It is of great significance for the smooth economic operation of the multi-MGs.This paper first introduces the structure and operation control mode of the MG,and establishes the mathematical model and economic cost optimization model of each distributed generation unit.Photovoltaic and wind turbines are used as representatives of renewable energy to analyze their characteristics.The battery is selected as the energy storage unit and the mathematical model is built.The diesel generator is used as the backup power source for analysis and modeling.When modeling the load,the load is classified according to the importance degree.It is involved in the economic dispatch of the MG and establishes a demand side response mechanism.Secondly,in order to solve the randomness and volatility of renewable energy generation in MG,a multi-time scale optimization method is introduced to optimize the MG from two time scales.The long-term scale optimization method optimizes the hourly load scheduling strategy for the next day based on the forecast information of the previous day and the minimum operating cost of the MG.The short-time scale is based on the long-term scale optimization results.Based on the short-term intraday forecast,the 15min-level rolling optimization is performed to correct the deviation of the long-term scale due to the prediction error,and the optimal output value of each distributed power source is calculated.Thirdly,the non-cooperative game model of MG group is constructed,and a new method of multi-MG energy trading is proposed.The MG is divided into consumers and providers based on the balance between internal power generation and its own load.Providers provide energy based on the proportion of consumers' bids and benefit from the proportion of energy they provide.Each consumer maximizes its own interests by competing with other consumers by changing the bid to get the energy provided by the provider.At the same time,the MG needs to consider the overall power balance and the constraints of the internal units in the bidding process.Furthermore,it is proved that the non-cooperative model belongs to the potential game,so there is Nash equilibrium and iteratively finds the Nash equilibrium solution through the particle swarm optimization algorithm.Finally,in order to prove the feasibility of the above optimization method and the superiority of the optimization effect,the single MG and the multi-MG including three MGs are configured by HOMER software.On this basis,the multi-time scale optimization method of single MG and the non-cooperative game energy trading method of multi-MG are simulated,and the research on MG energy optimization scheduling is summarized.