改进的带有局部搜索算子的量子粒子群算法  

Improvement of quantum-behaved particle swarm optimization with generalized local search operator

作　　者：蔡继亮[1] 叶微[1] 

机构地区：[1]空军工程大学理学院,陕西西安710051

出　　处：《计算机工程与设计》2010年第14期3264-3267,共4页Computer Engineering and Design

基　　金：国家自然科学基金项目(60573040)

摘　　要：带有局部搜索算子的量子粒子群算法(MQPSO-LQPSO)是一种较成功的改进的QPSO算法,但是该算法在搜索震荡的不足,在一定程度上降低了搜索效率。针对该问题,提出了一种改进方法,将LQPSO搜索得到的最优粒子替换MQPSO的Gbest和当前群中适应度最佳的粒子和最差的粒子。在标准测试函数上的仿真实验结果表明,改进的算法在不改变原有算法框架和不引入新的参数条件下,提高了MQPSO-LQPSO的搜索能力和计算效率。Quantum-behaved particle swarm optimization with generalized local search operator （shortly,MQPSO-LQPSO） is a very successful algorithm of modified QPSO proposed. However,problem such as concussive search lowered its searching efficiency. Correspondingly,an improved MQPS-LQPSO is proposed. The best particle of LQPSO is sent to the best particle and the worst particle in the swarm as well as the Gbest of the current MQPSO. The test on benchmark functions show that the improved algorithm improves the searching ability and raises computational efficiency without changing the basic frame of MQPSO-LQPSO or adopting any new parameters.

关 键 词：量子粒子群算法 局部搜索算子 搜索震荡 MQPSO当前群中最佳粒子 搜索效率 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]