摘要
介绍了一种易于实现、参数少且收敛快的集群智能算法—微粒群算法,并将其应用于梯级水电厂的短期优化调度。提出以确定微粒群在多维空间中的最优位置来实现多阶段优化调度决策的方法,并针对算法易陷入局部最优的缺陷,引入遗传算法中的“杂交”因子以及采用自适应的惯性权重,以改进其全局优化能力。通过实际算例验证了该算法的有效性,从而为梯级水电厂的短期优化调度问题提供了一种新的求解途径。
A swarm-intelligence-based algorithm of particle swarm optimization, which is simply implementing, fast convergent and only with few parameters, is introduced and applied to the short-term operation optimization of cascade hydropower plants. A method of finding the best location in multi-dimenslonal space of particles is presented to achieve the optimal decision of multi-stage operation, the global convergence performance of PSO is improved by importing a cross operator of GA and using a self-adapting inertia. The effectiveness of this algorithm is verified by the sample application, thus a new method is provided for the short-term optimization problem of cascade hydropower stations.
出处
《水力发电学报》
EI
CSCD
北大核心
2006年第2期94-98,共5页
Journal of Hydroelectric Engineering
基金
国家自然科学基金资助项目(50579019)
关键词
水利管理
短期优化调度
微粒群算法
梯级水电厂
water management
shor-term operation optimization
particle swarm optimization （PSO）
cascade hydropower stations
