摘要
以某电厂自并励励磁系统现场实测的阶跃响应曲线为例,介绍了在自并励励磁系统参数辨识中,如何选择模型结构,如何借助于MATLAB中的系统辨识工具箱辨识模型参数。在列写待辨识励磁系统的状态方程时,为了避免对阶跃函数求导,可将励磁系统传递函数框图的结构进行调整,并适当选择状态变量。文中给出了辨识结果,并将相同激励下,辨识模型的仿真曲线与实测曲线进行对比,二者拟合很好,说明所提出的辨识方法是可行的;最后提出了励磁系统参数辨识的注意事项:所加阶跃扰动的幅值不宜过大,否则会导致限幅环节动作,大大增加了非线性因素,这样就无法用所测数据来辨识线性环节的参数;在现场试验中,如果能够增加测点,分环节辨识,辨识结果将更加准确。
Taking the step response curves of a bus fed static excitation system in a power plant as an example, this paper investigates how to select the model structure for the identification of bus fed static excitation system parameters, and how to identify the parameters using the System Identification Toolbox in MATLAB. In order to avoid calculating the derivative of the step function, the structure of the block diagram of excitation system＇s transfer function can be modified, and the state variables should be properly selected. The simulation curves of the identified models fit the test curves well. Based on the experience on the identification of excitation system parameters, this paper also gives two remarks for guiding the identification procedure： the amplitude of the step can＇t be too large, otherwise the limiters will operate and the nonlinear factor increases greatly, then the measured data can＇t be used to identify the parameters of the linear parts; if the measuring points can be increased and every part can be identified separately, the results of identification will be more accurate.
出处
《现代电力》
2005年第6期32-34,共3页
Modern Electric Power
关键词
自并励励磁系统
参数辨识
模型结构
MATLAB
系统辨识工具箱
bus fed static excitation system
parameters identification
model structures MATLAB
system identification toolbox
