An evolutionary membrane algorithm for global numerical optimization problems

Nature-inspired algorithms for optimization are significant topics in the areas of computational intelligence. The contribution of this paper is to present a new heuristic intelligent evolutionary algorithm based on membrane systems to solve the global numerical optimization problems. The proposed algorithm employs the fundamental ingredients of membrane systems, including multisets, reaction rules and membrane structure. In addition, the proposed algorithm incorporates information of the adjacent symbol-objects, to guide the evolution toward the global optimum, efficiently. More specifically, symbol-objects are evolved by the cellular automata model which invokes the rewrite rules to exchange the information of the adjacent symbol-objects. Moreover, sharing information in the skin membrane is implemented, which accelerates the speed of the proposed algorithm to find the global optimal solution. In the extensive experimental study, the effectiveness of the proposed algorithm is demonstrated with the benchmark global numeric optimization problems. The experimental results indicate that the proposed method is a competitive optimizer in comparison with the four state-of-the-art evolutionary algorithms.