【摘要】 针对量子遗传算法(QGA)易陷入局部极值、具有早熟收敛等问题,分析了QGA的流程,从全局搜索和局部搜索两个层面探讨了QGA的改进策略,提出了一种新的算法。该算法利用混沌运动的遍历性和随机性进行全局搜索,同时利用梯度信息对QGA的量子更新过程环节进行优化。典型函数测试分析表明,该方法的综合性能明显优于量子遗传算法及遗传算法。更多还原

【Abstract】 Aiming at the trouble of easy getting into local minimum and premature convergency existed in quantum genetic algorithm,this paper analysed the flow of QGA,improved the strategy in two sides of global searching and local searching,and presented a new algorithm.This algorithm executed global search using the chaos movement’s ergodicity and randomness,in the same time optimized the renovation process of quantum with the gradient information.The test of typical function shows that the performance of this kind of method is better than quantum genetic algorithm and genetic algorithm.更多还原