Evaluation of Combined Pareto Multiobjective Differential Evolution on Tuneable Problems


Abstract

Many optimization problems in engineering involve the satisfaction of multiple objectives within the limits of certain constraints. Methods of evolutionary multi-objective algorithms (EMOAs) have been proposed and applied to solve such problems. Recently, a combined Pareto multi-objective differential evolution (CPMDE) algorithm was proposed. The algorithm combines Pareto selection procedures for multi-objective differential evolution to implement a novel selection scheme. The ability of CPMDE in solving unconstrained, constrained and real optimization problems was demonstrated and competitive results obtained from the application of CPMDE suggest that it is a good alternative for solving multi-objective optimization problems. In this work, CPMDE is further tested using tuneable multi-objective test problems and applied to solve a real world engineering design problem. Results obtained herein further corroborate the efficacy of CPMDE in multi-objective optimization.