A weighted corrective fuzzy reasoning spiking neural P system for fault diagnosis in power systems with variable topologies. (June 2020)

• Record Type: Journal Article
• Authors: Wang, Tao; Wei, Xiaoguang; Wang, Jun; Huang, Tao; Peng, Hong; Song, Xiaoxiao; Cabrera, Luis Valencia; Pérez-Jiménez, Mario J.
• Is Part Of: Engineering applications of artificial intelligence. Volume 92(2020)
• Publication Date: 2020-06
• Subjects: Fault diagnosis -- Power system -- Spiking neural P system -- Fuzzy reasoning -- Membrane computing -- Cause–effect network; Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2020.103680
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: This paper focuses on power system fault diagnosis based on Weighted Corrective Fuzzy Reasoning Spiking Neural P Systems with real numbers (rWCFRSNPSs) to propose a graphic fault diagnosis method, called FD-WCFRSNPS. In the FD-WCFRSNPS, an rWCFRSNPS is proposed to model the logical relationships between faults and potential warning messages triggered by the corresponding protective devices. In addition, a matrix-based reasoning algorithm for the rWCFRSNPS is devised to reason about the fault alarm messages using parallel representations. Besides, a layered modeling method based on rWCFRSNPSs is developed to adapt to topological changes in power systems and a Temporal Order Information Processing Method based on Cause–Effect Networks is designed to correct fault alarm messages before the fault reasoning. Finally, in a case study considering a local subsystem of a 220kV power system, the diagnosis results of five test cases prove that the proposed FD-WCFRSNPS is viable and effective. Highlights: We propose Weighted Corrective Fuzzy Reasoning Spiking Neural P Systems with real numbers (rWCFRSNPSs) to form a new graphic fault diagnosis method of power systems. We develop a layered modeling way for rWCFRSNPS-based fault diagnosis models to adapt to topological changes in power systems. We design a Temporal Order Information Processing Method based on Cause–Effect Networks to correct fault alarm messages.