摘要
大力开发水电资源是解决我国能源供应紧张问题的重要途径。水轮机辨识多采用基于线性系统辨识理论的算法,由于缺乏对水轮机非线性因素的考虑,很难满足电力系统实际运行状况分析的需要。智能优化算法是水轮机辨识的最优斱法,本文将差分进化算法和粒子群算法成功应用于水轮机辨识中,幵针对智能优化算法存在的原理误差,提出将两种算法辨识得到的平均值作为辨识结果。仿真结果显示该策略辨识出的系统参数精确度高,误差小。
Developing hydro power resources vigorously is an important way to solve China＇s energy supply tension. The algorithm based on the theory of linear system identification for the most part is adopted to identify hydraulic turbine. Due to lacking of consideration of nonlinear factors of hydro-turbine, it is difficult to meet the actual operation of the electric power system analysis. Intelligent optimization algorithm is a preferred method of identification turbine. In this paper, the differential evolution and the particle swarm optimization are applied to the hydro-turbine identification. According to the error in principle existing in intelligent optimization algorithms, the method that the average value of two algorithms as the identification results presented, the simulation results show that the transfer function identify from this strategy is precise, and less system error.
出处
《大电机技术》
2016年第3期58-61,64,共5页
Large Electric Machine and Hydraulic Turbine
关键词
粒子群算法
差分进化算法
水轮机调速系统
参数辨识
非线性系统
particle swarm optimization algorithm
differential evolution algorithm
hydro turbinegoverning system
parameter identification
nonlinear system
