Quantum-Inspired Immune Clonal Multiobjective Optimization Algorithm

摘要

Based on the concept and principles of quantum computing, a quantum-inspired immune clonal multiobjective optimization algorithm (QICMOA) is proposed to solve extended 0/1 knapsack problems. In QICMOA, we select less-crowded Pareto-optimal individuals to perform cloning, recombination update. Meanwhile, the Pareto-optimal individual is proliferated and divided into a set of subpopulation groups. Individual in a subpopulation group is represented by multi-state gene quantum bits. For the novel representation, qubit individuals in subpopulation are updated by applying a new chaos update strategy. The proposed recombination realizes the information communication among individuals so as to improve the search efficiency. We compare QICMOA with SPEA, NSGA, VEGA and NPGA in solving nine 0/1 knapsack problems. The statistical results show that QICMOA has a good performance in converging to true Pareto-optimal fronts with a good distribution.