Research On Energy Management Problems For Microgrid Based On Robust Optimization

The Energy Internet,which contains a plurality of microgrids,various types of independent power generators and users,is one of the important means to solve the current energy crisis and environment problems,and it is also called as the key of the Third Industrial Revolution.In this paper,the energy management problem for microgrids,which is an important unit of the Energy Internet,is mainly studied.The study considers various uncertainties in microgrids and uses a robust optimization method to deal with these uncertainties.A serious of energy management problems for two types of microgrids are modeled,analyzed and discussed,in which electric vehicles,a special type of loads,have also been considered at last.The main work can be concluded as follows.Firstly,an energy optimization scheduling problem for isolated microgrids is studied.By using the mixed logic dynamic model and considering the source,storage,load units of microgrids,a mixed integer programming model for microgrid energy management is established.Then,considering the uncertainty and dynamics of the output of the load and the renewable energy in microgrids,we proposes a robust model predictive control strategy for energy optimal scheduling.Under this strategy,by controlling the controllable power generators,the energy storage and the flexible load,the real-time on-line energy management and the reasonable allocation of the power for the microgrid system are realized.The proposed strategy can effectively reduce the negative impact of uncertainties on microgrid energy management and scheduling,and exhibits advantages compared with the traditional model predictive control strategy.Secondly,an energy management problem for the microgrids connected to the grid is discussed.Considering both the impact on the grid and the economic operation during the microgrid operation process,a multi-time scale robust optimization scheduling approach for microgrid energy management is established.This approach combines the day-ahead scheduling strategy and real-time optimal energy dispatch,and the operational objects include the controllable power generators,the energy storage,the flexible load and the interactive power with the grid.The robust optimization model is compared with the deterministic one,which reveals that the robust one is better in economy and security.At last,the influence of robust optimization parameter settings on energy management scheduling results is analyzed.Finally,under the background of an actual microgrid,an improved multi-time scale robust optimization scheduling strategy is proposed,considering the optimal scheduling of electric vehicles.By adjusting the charging and discharging power of the energy storage system,the interaction plan with the grid,and the charging time and power of the electric vehicle,the global economic dispatch of the energy for the microgrid is realized.Compared with the deterministic modeling method,it is proved that the proposed model has a better inhibitory effect considering the system uncertainties.This research has good theoretical and practical value for the energy management of both the isolated and grid-connected microgrids.