A data-driven approach for fault time determination and fault area location using random matrix theory. (March 2020)

• Record Type: Journal Article
• Authors: Xiong, Yongxin; Yao, Wei; Chen, Weibiao; Fang, Jiakun; Ai, Xiaomeng; Wen, Jinyu
• Is Part Of: International journal of electrical power & energy systems. Volume 116(2020)
• Publication Date: 2020-03
• Subjects: Data-driven approach -- Phasor measurement unit (PMU) -- Wide-area measurement system (WAMS) -- Fault diagnosis -- Random matrix theory (RMT) -- Bad data; Electrical engineering -- Periodicals; Electric power systems -- Periodicals; Électrotechnique -- Périodiques; Réseaux électriques (Énergie) -- Périodiques; Electric power systems; Electrical engineering; Periodicals; 621.3
• DOI: 10.1016/j.ijepes.2019.105566
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: This paper proposes a wide-area measurement-based data-driven approach for fault time determination and fault area location to avoid the influence of bad data. The proposed approach has no requirement for impedance information compared with the existing impedance-based methods. The proposed approach is verified via not only IEEE-39 bus system but also a practical provincial power grid of China. An assistant software is developed using the proposed method to help dispatchers obtain fault information more accurately. Abstract: This paper proposes a wide-area measurement-based data-driven approach for fault time determination and fault area location in the power system. To avoid the influence of bad data, the random matrix theory (RMT) is applied to the proposed approach. The measurements obtained from wide-area measurement system (WAMS) are used to form the raw data matrix, its mean spectral radius (MSR) index and the correlation among entries are employed to determine the fault time and locate the fault area using RMT. Since the proposed approach only requires the data obtained from WAMS and is a data-driven method, no physical model and topology information is needed compared with the existing model-based methods. Case studies are carried out on the IEEE 39-bus power system and a practical provincial power grid of China respectively, and a decision-support tool is programmed using the proposed method to help operators obtain fault information more timely and accurately. Simulation results show that the proposed method can determine the fault time precisely and locate the fault line, whether there is bad data in wide-area measurements or not.