Improved Strategies of Multi-objective Differential Evolution (MODE) for Multi-objective Optimization
Ashish M. Gujarathi, B. V. Satish Babu · 2011
Abstract. Multi-objective optimization using an evolutionary computation technique is used extensively for solving conflicting multi-objective optimization problems. In this work, an improved strategy of multi-objective differential evolution (MODE) where the mutation strategy is changed to a trigonometric mutation approach is proposed. The proposed strategy along with other well known strategies of MODE is used to compare the performance metrics (such as convergence and divergence) with other evolutionary algorithms from the literature. The Pareto optimal solutions are obtained for benchmark test functions and are compared using several strategies of MODE. Improved strategies of MODE show a competitive performance when compared with other evolutionary multi-objective optimization algorithms (EMOAs). Keywords: Multi-objective optimization; evolutionary algorithms, trigonometric mutation; differential evolution; optimization 1