Study On Robust Multi-objective Optimization For Microgrid Power Dispatch Under Uncertain Environment

Microgrid refers to a small power system composed of various distributed power sources,loads,energy storage devices,and control equipment.When renewable energy such as photovoltaics and wind power exists,the microgrid system may become more complex due to the uncertainty and dynamics of these sources.Effectively coordinating and controlling the output of each distributed power source in a microgrid is of utmost importance as it ensures the safe,stable,economical,and reliable operation of the power system.Microgrid power dispatch typically involves multiple aspects such as economy,reliability,and environmental protection.These dispatch objectives often conflict with each other,and the dispatch plan must strike a balance and make trade-offs between different objectives.Moreover,the uncertainty of renewable energy output may lead to a deterioration in the performance of dispatch schemes.When the microgrid contains a large number of distributed power sources,the uncertainty dimension of the model increases,as well as the algorithm complexity of robust optimization.To address the above issues,this paper focuses primarily on the robust multiobjective optimization problem of microgrid power dispatch.First,a model is constructed for the microgrid power dispatch optimization problem.Second,a method is designed based on reference vector and domination robustness.Finally,a scheme is proposed built on robust estimation and dual evolution.These measures effectively enhance the efficiency of robust optimization and realize power dispatch of microgrid under uncertain environment.The specific work is as follows.1.The microgrid power dispatch is modeled.The optimization objectives are the economic benefits,environmental benefits,and renewable energy penetration rate of the microgrid system under uncertain conditions.Based on the renewable energy output prediction interval,Uncertainty of distributed power output under power balance constraints is analyzed,and a microgrid day-ahead optimization dispatch model is constructed.2.A robust optimization method for the microgrid optimization dispatch model is designed.This method(DR-MOEA)is based on reference vector and dominance robustness,which mainly includes a new robustness evaluation index to evaluate the robustness of individuals by using the concept of robust dominance combined with the relative fluctuation distance between sampled individuals.Moreover,a selection strategy based on reference vectors is used to maintain the diversity of solutions,and the solutions with good robustness are reserved.The experiment issues are benchmark problems and microgrid power dispatch models,and the results show that DR-MOEA can effectively maintain the robustness and diversity of the population while satisfying convergence.3.An algorithm based on robust estimation and dual evolution is proposed(RMOEA-REDE).This algorithm focuses on the challenge of high computational complexity in robust optimization.The decision variable classification strategy is adopted to reduce model complexity by analyzing low-sensitivity robust-related variables and high-sensitivity robust-related variables respectively.In addition,a candidate solution evaluation strategy based on robust Rank and convergence Rank,and a selection strategy based on reference vectors are proposed to improve the convergence speed of the algorithm while ensuring the robustness of the solutions.The experimental results show that RMOEA-REDE can efficiently obtain solutions with good robustness and optimality,maintain the convergence of the solution set,improve the robustness of the solution set,and exhibits stronger overall performance.