New Evolutionary Algorithm to Solve Dynamic Constrained Optimization

Abstract

In this paper, a new multiobjective evolutionary algorithm for dynamic nonlinear constrained optimization problems (DNCOPs) is proposed. First, the time period is divided into several equal subperiods. In each subperiod, the DNCOPs is approximated by a static nonlinear constrained optimization problem (SNCOPs). Second, for the SNCOPs, inspired from the ideal of multiobjective optimization, it is transformed into a static bi-objective optimization problem. Third, a new multi- objective evolutionary algorithm (DMEA) for DNCOPs is proposed and the simu- lation results indicate the proposed algorithm is effectiveness for dynamic nonlinear constrained optimization problems.