An Effective Dynamical Multi-objective Evolutionary Algorithm for Solving Optimization Problems with High Dimensional Objective Space

摘要

An effective dynamical multi-objective evolutionary algorithm (DMOEA) based on the principle of the minimal free energy in thermodynamics was proposed in the paper. It provided a new fitness assignment strategy based on the principle of free energy minimization of thermodynamics for the convergence of solves, introduced a density-estimate technique for evaluating the crowding distance between individuals and a new criterion for selection of new individuals to maintain the diversity of the population. By using multi-crossover operator, it improved the search efficiency and the robustness. The test example results proves the validity of the algorithm in its rapidly convergence and maintaining diversity.