基于杂交粒子群算法的汽轮机调速系统参数辨识

Parameters Identification for Steam Turbine Governing System based on Hybrid Particle Swarm Optimization

汽轮机调速系统对于机组动态行为及电力系统的电能质量有重要的影响,准确模拟调速系统运行控制特性对于提高电力系统稳定性水平意义重大。以BPA软件中的调速系统标准模型作为待辨识模型,通过对汽轮机施加频率小阶跃扰动,选择对频率阶跃响应影响较大的参数作为待辨识参数,采用杂交粒子群算法分别对调速器、执行机构及汽轮机环节进行辨识。在Simulink中搭建单机-无穷大系统,采用辨识模型计算得到的仿真值与与实测值的误差较小,表明将杂交粒子群算法应用于调速系统参数辨识的方法是有效的。

Generator governing system has an important effect on the unit dynamic behavior and power quality of electric system. The accurate simulation of generator governing system is of great significance to the improvement of stability of the electric system. In this paper, I take the governing system＇s standard model in BPA software as the identification model, by applying small frequency step to the turbine to select the parameters that have greater influence by the frequency step as the identification parameters, and then use the hybrid particle swarm algorithm to identify the governor, actuator and turbine link respectively. Finally, I construct the single-infinite power system in Simulink to check the identification parameters and come to the conclusion that the simulation curves are similar to the experimental results. This indicates that the method to use the hybrid particle swarm algorithm to identify the governing system parameters is developed.