Improved efficiency of MOPSO with adaptive inertia weight and dynamic search space
Lee-Ping Pang, Sin-Chun Ng · Proceedings of the Genetic and Evolutionary Computation Conference Companion · 2018
In this paper, a new Multi-Objective Particle Swarm optimization algorithm (MOPSO) with adaptive inertia weight and dynamic search space is introduced for multi-objective optimization. The objective of the study is to investigate an efficient MOPSO to deal with large-scale optimization and multi-modal problems. The new adaptive inertia weight strategy allows the inertia weight to keep varying throughout the algorithm process, which helps the algorithm to escape from local optima. The dynamic search space design can avoid decision variables from continuously taking their extreme values, and therefore enhances the searching efficiency. The performance of the proposed algorithm was compared with three popular multi-objective algorithms in solving seven benchmark test functions. Results show that the new algorithm can produce reasonably good approximations of the Pareto front, while performing with a budget of 10,000 fitness function evaluations.