A hybrid genetic algorithm for unconstrained global numerical optimisation

This paper presents a hybrid genetic algorithm for unconstrained global numerical optimisation. The proposed approach in this paper considers a sampling procedure based on orthogonal design and quantisation technology, the use of an orthogonal genetic algorithm with quantisation for global exploration, and the application of a local optimisation technique for local exploitation. This proposed new approach is applied to 10 multi-modal problems. A comparative study focuses on the overall search effectiveness in terms of the local minima found and required function evaluations. The results obtained from the computational example have shown that the proposed algorithm is correct, feasible and effective.