Bayesian multiple kernels learning-based transient stability assessment of power systems using synchronized measurements

Abstract

A new method for transient stability assessment (Tsa) of power systems using Bayesian multiple kernels learning and synchronized measurements is presented in this paper. The proposed scheme extracted the initial features symbolizing the stability of power systems from synchronized measurements and broke the features into three subsets: the features immediately following a fault, the features at the fault clearing time and the features after the fault clearing time, then instructively combined feature spaces corresponding to each feature subset through Bayesian multiple kernels learning and finally determined the transient stability based on the trained TSA model. The novelty of the proposed method is in the fact that it improves the classification accuracy and reliability of TSA by combining different feature spaces. The test results on the New England 39-bus test system verify the validity of the presented method.