Machine Learning-Aided Generator Coherency for Hybrid Power System Stability Enhancement

View at publisher Summarise with VIRAT
Authors: Sudipta Ghosh, Abdula Abu Zuhri, Davor Svetinović, Hany Mohamed Hasanien, Surour Alaraif, Mohamed Shawky El Moursi
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This paper presents a novel method for determining the most effective generator coherency groupings to control damping of inter-area oscillation modes in electrical grids. The research proposes an advanced generator coherency identification technique utilizing machine learning (ML) to enhance damping through model predictive control (MPC) in a wide-area damping controller (WADC). The innovative coherency identification approach employs hierarchical clustering based on four criteria: minimum distance, minimum phase shift, maximum overlapping, and link inconsistency. The WADC framework incorporates several techniques, including modal analysis, residual analysis, parameter estimation, state estimation, and MPC for the damping controller. Additionally, Kalman Filters are employed to update the controller's states and eliminate noise from the signals measured by strategically placed phase measurement units (PMUs). The proposed method's effectiveness is evaluated using the Wind integrated IEEE 39 bus and IEEE 68 bus test system. The outcomes show superior damping performance and precision across all three coherency grouping performance metrics when compared to two recently published algorithms: SVD K-means-based clustering and adjacency matrix-based clustering techniques.

Fields of Study & Subjects

Power System Optimization and Stability Frequency Control in Power Systems Microgrid Control and Optimization

Publication details

Journal / source
IEEE Transactions on Industry Applications
Pages
1–14
Publication date
1 January 2026
Language
EN