PROJECT TITLE :

Online Damping Ratio Prediction Using Locally Weighted Linear Regression

ABSTRACT:

During this study, a regionally weighted linear regression (LWLR) methodology is proposed to predict damping ratio of a dominant mode online. The LWLR methodology, that is nonparametric and knowledge-oriented, is basically proposed for nonlinear information fitting; so, it will track the nonlinear power system operations and help damping ratio prediction in real power systems, which is hardly achieved by the conventional linear regression. To successfully implement this methodology, the measurement of weighting value and the choice of weighting operate in addition to its parameter setting, connected to prediction accuracy and numerical conditions, are extensively mentioned. Simulations are dole out in a very two-area four-machine system and a massive complex system, China Southern Grid. Both results validate the effectiveness of the proposed method.