摘要
针对基本粒子群算法在处理高维复杂问题时易陷入局部极值点的不足,提出自适应随机惯性权的粒子群优化算法,并基于此算法建立起洪水过程的放大模型。在进化过程中,为粒子群中的各个粒子随机选取惯性权值,并随进化代数自适应地调整随机惯性权值的概率分布,提高了算法全局寻优的性能。实例表明:该模型不但可以有效保持典型洪水的模式,避免手工修均的任意性,而且可以很好地满足洪峰洪量的约束要求,有较高的精度。
Aiming at the deficiency of the traditional particle swarm Optimization, the new zoom model calculated by particle swarm optimization based on adaptive random inertia weight is put forward. In the process of evolution, the weight of every particle is selected randomly and the probability density function of random inertia weight is adjusted with the times of evolution. Examples indicate that the model is convenient, accurate and can satisfy the constraints of actual peak flood discharge and flood volume keeping the mode of the typical flood effectively.
出处
《中国农村水利水电》
北大核心
2008年第5期1-4,7,共5页
China Rural Water and Hydropower
基金
河南省高校创新人才培养工程项目(HNCX2003-17)
关键词
设计洪水
粒子群
随机惯性权
design flood
particle swarm
random inertia weight
