Identification of generator criticality and transient instability by supervising real-time rotor angle trajectories employing RBFNN

Bhanu Pratap Soni; Akash Saxena; Vikas Gupta; S L Surana

doi:10.1016/j.isatra.2018.08.008

Identification of generator criticality and transient instability by supervising real-time rotor angle trajectories employing RBFNN

ISA Trans. 2018 Dec:83:66-88. doi: 10.1016/j.isatra.2018.08.008. Epub 2018 Aug 14.

Authors

Bhanu Pratap Soni¹, Akash Saxena², Vikas Gupta³, S L Surana²

Affiliations

¹ Department of Electrical Engineering, Malaviya National Institute of Technology, Jaipur, 302017, India. Electronic address: er.bpsoni2011@gmail.com.
² Department of Electrical Engineering, Swami Keshvanand Institute of Technology, Management & Gramothan, Jaipur, 302017, India.
³ Department of Electrical Engineering, Malaviya National Institute of Technology, Jaipur, 302017, India.

PMID: 30209021
DOI: 10.1016/j.isatra.2018.08.008

Abstract

Identification of transient stability state in real-time and maintaining stability through preventive control technology are challenging tasks for a large power system while integrating deregulation constraints. Widely employment of the phasor measurement units (PMUs) in a power system and development of wide area management systems (WAMS) give relaxation to monitoring, measurement and control hurdles. This paper focuses on two research objectives; the first is transient stability assessment (TSA) and second is selection of the appropriate member for the control operation in unstable operating scenario. A model based on the artificial machine learning and PMU data is constructed for achieving both the objectives. This model works through prompt TSA status with radial basis function neural network (RBFNN) and validates it with PMU data to determine the criticality level of the generators. To reduce the complexity of the model a transient stability index (TSI) is proposed in this paper. A RBFNN is used to determine the transient stability aspects like stability status of system, coherent group and criticality rank of generator and preventive control action, following a large perturbation. PMUs measure post-fault rotor angle values and these are used as input for training RBFNN. The proposed approach is demonstrated on the IEEE 10-generator 39-bus, 16-generator 68-bus and 50-generator 145-bus test power systems successfully and the effectiveness of the approaches is discussed.

Keywords: Coherency identification; Dynamic stability assessment; PMU; Preventive control action; RBFNN; Rotor angle stability; Transient instability detection & control.