Ensemble Differential Evolution with dynamic subpopulations and adaptive clearing for solving dynamic optimization problems

Many real-life optimization problems are dynamic in time, demanding optimization algorithms to perform search for the best solutions in a time-varying problem space. Among population-based Evolutionary Algorithms (EAs), Differential Evolution (DE) is a simple but highly effective method that has been successfully applied to a wide variety of problems. We propose a technique to solve dynamic optimization problems (DOPs) using a multi-population version of DE that incorporates an ensemble of adaptive mutation strategies with a greedy tournament global search method, as well as keeps track of past good solutions in an archive with adaptive clearing to enhance population diversity.