HypDE: A hyper-heuristic based on differential evolution for solving constrained optimization problems

Abstract

In this paper, we present a hyper-heuristic, based on Differential Evolution, for solving constrained optimization problems. Differential Evolution has been found to be a very effective and efficient optimization algorithm for continuous search spaces, which motivated us to adopt it as our search engine for dealing with constrained optimization problems. In our proposed hyper-heuristic, we adopt twelve differential evolution models for our low-level heuristic.We also adopt four selection mechanisms for choosing the low-level heuristic. The proposed approach is validated using a well-known benchmark for constrained evolutionary optimization. Results are compared with respect to those obtained by a state-of-theart constrained differential evolution algorithm (CDE) and another hyper-heuristic that adopts a random descent selection mechanism. Our results indicate that our proposed approach is a viable alternative for dealing with constrained optimization problems. © Springer-Verlag Berlin Heidelberg 2013.