A real-coded quantum clone multi-objective evolutionary algorithm

In this paper, a real-coded quantum clone multi-objective evolutionary algorithm(RQC-MOEA) is proposed for multi-objective optimization. Based on the concepts of quantum computation, a triploid chromosome is constructed by the real-coded decision vector and the chaos encoding probability amplitude. The Gaussian mutation is implemented on the decision variables to increase the diversity of population by using probability amplitude α, β as standard deviation, then the probability amplitude is updated by the quantum rotation gate, this process helps RQC-MOEA achieve a balance between exploration and exploitation. Also clone is utilized to accelerate the convergence speed and the improved crowding-distance makes the algorithm better distributed. Compared with NSGA2 and SPEA2, RQC-MOEA possesses a good balance among convergence, distribution, diversity and execution time. Especially for the Pareto non-dominated set is relatively complicated or difficult convergence optimization problems, RQC-MOEA can get a more widely spread Pareto front and has a clear advantage on convergence and distribution.