Evolutionary Large‐Scale Multi‐Objective Optimization in Power Systems

This chapter explores evolutionary algorithms applied to power systems, with a particular focus on a case study involving the ratio error estimation of voltage transformers (VTs). In the case study, the chapter discusses the importance of ratio error estimation of VTs and the limitations of existing methods. It introduces the time‐varying ratio error estimation (TREE) method as an innovative solution to address the shortcomings of existing methods. Next, the chapter presents a data‐driven approach for formulating the TREE problem as a large‐scale multi‐objective optimization, detailing the process of extracting statistical and physical principles from power distribution systems. Finally, a problem knowledge‐driven coevolutionary algorithm is designed to solve the TREE problems, demonstrating the potential of evolutionary algorithms in addressing complex real‐world problems in power systems. A novel genetic operator is proposed to improve the efficiency of generating sparse solutions, thus enabling the Pareto optimal solution sets of TREE problems to be solved effectively.