Disturbance Magnitude Estimation: MLP-based Fusion Approach for Bulk Power Systems

Abstract

Power system disturbances can damage electrical components or even collapse an interconnected power grid. The accurate estimation for the disturbance magnitude is critical in ensuring the reliability of the power grid and protecting electrical components. To address this issue, this paper proposes a machine learning approach to estimate the disturbance magnitude. This approach combines the estimations of the conventional approaches to provide a more accurate estimation. Evaluated with the confirmed cases in western interconnection and field-collected measurements from FNET/GridEye, the proposed method achieves 91.2% accuracy on magnitude estimation, which is 7% better than the conventional approaches. Moreover, the proposed method does not require a complex system topology, which makes it adaptive to various sizes of power systems.