一种求解多目标数值优化问题的实值量子演化算法

基于多目标优化原理和量子计算原理，提出了一种求解多目标数值优化问题的实值量子演化算法. 该算法除保留求解单目标优化问题的实值量子演化算法的特点外，还有三个主要特征:首先，根据多目标优化特点,使用多目标密度比较算子对种群进行排序和筛选；其次，应用非均匀变异算子保持解的收敛性和提高局部搜索能力；再次，使用多样性保持算子来保持解的多样性. 实验中使用多目标优化测试集并且同NSGA-II算法相比较，结果表明了算法的有效性.

A Real-Valued Quantum-Inspired Evolutionary Algorithm for Solving Multi-Objective Numerical Optimization Problems

Based on the principles of quantum computing and multi-objective optimization, a real-valued quantum-inspired evolutionary algorithm(MO-RQIEA) for solving multi-objective numberical optimization problems is proposed. Except that the specifications of the single-objective real-valued quantum-inspired evolutionary algorithm(RQIEA) are kept, the proposed algorithm has three main features to enhance the RQIEA. First, the crowed comparison operator is used to sort and select individuals in multi-objective sense. Second, the non-uniform mutation operator is applied to improve the precision for local search and to preserve the convergency for the solutions. Third, the diversity preserve operator is proposed to maintain the diversity. The proposed algorithm has been evaluated with multi-objective optimization test functions and compared against NSGA-II. The experiment results show the effectiveness of the proposed algorithm.