在众多研究领域都存在着客观或者人为的不确定优化问题，传统方法很难解决此类问题。论文在简述了传统量子遗传算法的原理和结构的基础上，分析了传统量子遗传算法主要存在的问题，即解空间转换和如何确定量子门的旋转相位，以此进行算法的改进，给出了改进量子遗传算法的流程，并以 Shaffer’s F1多峰不确定优化问题为例，分析了 IQGA的运行效率、收敛速度等性能。通过仿真研究表明 IQGA运行效率较高，收敛速度较快，能较好地支持不确定规划问题。%There is an obj ective or artificial uncertain optimization problem in many area,the traditional methods are difficult to solve such problems.The paper firstly introduces the principle and structure of the traditional quantum genetic al-gorithm (QGA),analyzes the main problem of the traditional quantum genetic algorithm,namely the problem of the solution space conversion,and how to determine the rotational phase of the quantum gate.Then the paper improves the algorithm based on the analysis,gives the process of improved quantum genetic algorithm (IQGA),and takes Shaffer's F1 multimodal uncertain planning for example,analyzes the properties of the running efficiency and the convergence efficiency etc.of IQ-GA.The simulation results show that the running efficiency of IQGA is higher,and convergence efficiency is faster,there-fore,the uncertain planning problem can be better supported by IQGA.
展开▼
