±20 Mvar STATCOM采用当地输入信号的阻尼控制器设计

DESIGN OF DAMPING CONTROLLER WITH LOCAL INPUT SIGNAL FOR±20 Mvar STATCOM

电力系统暂态稳定控制问题的难点是要解决一个具有强非线性、不确定性的大系统控制问题。在实际多机系统中暂态稳定问题多以多模振荡的形式出现 ,迭代自学习控制方法是一种有效的控制方法 ,该方法的关键是如何能实时在线地获得机电振荡主要频率的信息。文中用最小二乘法在线辨识用于补偿的戴维南等值阻抗 ,并利用 Prony分析法计算当前系统主要机电振荡频率。基于辨识结果在线修正控制器的参数 ,使控制系统能在很短的时间内适应系统的变化。数字仿真结果证实了由于在线识别参数 。

The difficulty of solving transient stability problem is to deal with the strongly nonlinear,uncertain and large- scale system. In practical multi- m achine systems,the transient stability problem is often accompanied with multi- m ode oscillations.Thus the iterative learning control is one of the effective methods to damp these oscillations.The essential point of this control is to obtain the information about the m ain electromechanical oscillation frequency from local signals on- line and in real- tim e.In this paper,on- line least square m ethod is applied to identify the Thevenin- equivalent im pedance used in com pensation,and Prony analysis method is used to compute the main electromechanical oscillation frequency of the system. Based on the results identified,the controller parameters are on- line modified accordingly,and the control system can adaptto the variations of the main system in a very short period. The perform ance improvem ents due to on- line parameter identification are shown via digital simulations.