【摘要】 介绍了一种基于集群智能的优化算法———微粒群算法,该算法具有实现简单、参数少且收敛快的特点。结合水电站厂内经济运行问题实际,提出通过确定微粒群在多维空间中的最优位置来实现机组间的负荷优化分配;针对算法易于局部收敛的缺点,引入了遗传算法的交叉算子来保持种群的多样性,并采用自适应惯性权重改善算法的解空间搜索能力。最后通过实际算例验证了算法的有效性,从而为水电站厂内经济运行问题提供了一种新的求解途径。更多还原

【Abstract】 A new optimization algorithm based on swarm intelligence—Particle Swarm Optimization(PSO) is introduced herein.It is characterized by simple implementation,fast convergence and few parameters.According to the practical problems from the in-plant economic operation of hydropower station,it is put forward that the optimized loading distribution among the generation units can be implemented by determining the best location of particle swarm in the multi-dimension space;in which a cross operator of GA is introduced into the normal PSO to maintain the diversity of the swarm and to avoid the premature convergence,and then the searching ability in solution space of the algorithm is improved with a self-adaptable inertia weight as well.The effectiveness of this modified PSO is verified by a practical application sample,therefore,a new method is provided for solving the problems from the in-plant economic operation of hydropower station.更多还原