An effective differential evolution with level comparison for constrained engineering design

Solving constrained engineering design problems via evolutionary algorithms has attracted increasing attention in the past decade. In this paper, a simple but effective differential evolution with level comparison (DELC) is proposed for constrained engineering design problems by applying the level comparison to convert the constrained optimization problem into an unconstrained one and using the differential evolution (DE) to perform a global search over the solution space. In addition, the mutation factor of DE is set to be a random number to enrich the search behavior, and the satisfaction level increases monotonously to gradually stress the feasibility. The comparison results between the DELC and five existing algorithms from the literature based on 13 widely used constrained benchmark functions show that the DELC is of better or competitive performance. Furthermore, the DELC is used to solve some typical engineering design problems. DELC is of superior searching quality on all the problems with fewer evaluation times than other algorithms. In addition, the effect of the increasing rate of satisfaction level on the performances of the DELC is investigated as well.