摘要
智能算法用于求解最优潮流(Optimal Power Flow, OPF)问题的研究由来已久,而没有免费午餐定理(No Free Lunch Theorem, NFL)已经证明没有适用于所有优化问题的超级算法,故近年来大量智能算法及其改进算法被用来求解OPF问题。本文使用粒子群算法,花朵授粉算法及一种改进花朵授粉算法,以两种IEEE30节点算例系统和东北电网500kV网架系统为算例,研究了两种智能算法的参数取值对其求解最优潮流的性能的影响,仿真结果表明不同的参数取值对算法求解OPF时性能的影响较大,在不同算例上最优参数的取值并不相同。另外,两种算法及一种改进算法在求解不同算例的OPF问题时性能排序不一致,说明这几种算法在求解不同算例时并无普适性。
Application of intelligent algorithm in the calculation of optimal power flow (OPF) problem is long-standing. No free lunch (NFL) theorem has proved that there are no super algorithms which is applicable to all optimization problems. So a large number of intelligent algorithms and their modified versions have been proposed to solve OPF problems in recent years. This paper uses particle swarm optimization (PSO), flower pollination algorithm (FPA) and an modified flower pollination algorithm (MFPA) to solve OPF problem of three cases, including two IEEE 30-bus systems and the 500 kV power grid system of Northeast. This paper studies the influence of the parameters of two intelligent algorithms on the performance of OPF problem, which shows that the performance of algorithms in solving OPF problem is greatly affected by their corresponding parameters. The optimal parameters are not the same for different examples. In addition, the performance ranking of algorithms used in this paper is not consistent in solving the OPF problems of different cases. These algorithms have no universality in solving different calculation examples.
作者
范航
李泽
栗然
周莹
张凡
FAN Hang;LI Ze;LI Ran;ZHOU Ying;ZHANG Fan(State Grid Shaoxing Electric power Co. Ltd,Shaoxing 312000, China;Department of ElectricPower Engineering of North China Electric Power University, Baoding 071003, China;NortheastChina Grid Company Limited Electric Power Control Center, Shenyang 110000, China;Northwest Electric Power Design Institute Co., Ltd., Xi’an 710000, China)
出处
《电力科学与工程》
2019年第7期42-48,共7页
Electric Power Science and Engineering
关键词
智能算法
NFL定理
最优潮流
算法参数
intelligent algorithm
NFL theorem
optimal power flow
algorithm parameter
