Jiang Zhengang

Solving constrained optimization problems by using covariance matrix adaptation evolutionary strategy with constraint handling methods

Numerous constraint handling techniques were proposed in the past to be used with evolutionary algorithms (EA). According to the no free lunch theorem, there is no single algorithm that can consistently outperform over all other algorithms for all types of problems and conditions. Depending on factors like feasibility ratio, multi-modality and problem specific characteristics, the exploration power of the chosen EA, different constraint handling techniques can be effective on different problems. The performance of Covariance Matrix adaptation Evolutionary Strategy (CMA-ES) has been studied for unconstrained optimization problems. But, there has not been much research work done for the constrained counterpart. Motivated by this observation, we studied the performance of CMA-ES with three different constraint handling techniques (CHT) present in the literature. We conducted experiment to test the algorithms performance with each technique separately on CEC-2010 benchmark problem sets. The relative performance of the algorithm with respect to the constraint handling techniques and comparison with state-of-the-art algorithm is presented in this paper.