Transient voltage stability assessment and margin calculation based on disturbance signal energy feature learning

• Published in: Frontiers in energy research Vol. 12
• Main Authors: Chen, Yan, Huang, Zirui, Du, Zhaobin, Zhong, Guoduan, Gao, Jiawei, Zhen, Hongyue
• Format: Journal Article
• Language: English
• Published: Frontiers Media S.A 02.10.2024
• Subjects: disturbance signal energy; ID3 decision tree algorithm; information gain; support vector machine; transient voltage stability assessment; wide area measurement system (WAMS)
• DOI: 10.3389/fenrg.2024.1479478
• ISSN: 2296-598X

With the increasing variation of the network topology and the high complexity of the processing measurement data, the transient voltage stability assessment of the new power system is facing significant challenges in low accuracy and high time costs. To address the shortcomings of the existing method and apply it to online assessment, this paper proposes an assessment method based on feature learning for disturbance signal energy (DSE) from bus voltages. Firstly, the relationship between DSE and system transient voltage stability is established, and the calculation of DSE from bus voltage time series is detailed. Subsequently, a transient voltage stability assessment method based on the ID3 Decision Tree algorithm and DSE is proposed. Finally, by employing the Support Vector Machine (SVM) to construct the optimal boundary in the feature space formed by the key buses, the transient voltage stability margin (TVSM) for specific scenarios is proposed. Simulation results on the IEEE 39-bus system demonstrate that the proposed method can rapidly and accurately assess the transient voltage stability of the system and calculate the stability margin, providing intuitive and interpretable results with high engineering application value.