Root-cause analysis of occurring alarms in thermal power plants based on Bayesian networks. (December 2018)

• Record Type: Journal Article
• Authors: Wang, Jiandong; Yang, Zijiang; Su, Jianjun; Zhao, Yan; Gao, Song; Pang, Xiangkun; Zhou, Donghua
• Is Part Of: International journal of electrical power & energy systems. Volume 103(2018)
• Page Start: 67
• Page End: 74
• Publication Date: 2018-12
• Subjects: Industrial alarm systems -- Root cause analysis -- Bayesian networks -- Thermal power plants; 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.2018.05.029
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: A method is proposed to analyze root causes of occurring alarms in power plants. Baysien network describes the relation between alarm and root-cause variables. Probability parameters in the Bayesian network are updated recursively. The proposed method can well deal with three challenges in practice. Industrial examples illustrate the effectiveness of the proposed method. Abstract: The paper proposes a method to analyze root causes of occurring alarms, for a special type of alarm variables in thermal power plants, where alarms are arisen from binary-valued root-cause variables. A Bayesian network with one child node and multiple parent nodes is used to describe the relationship between an alarm variable and root-cause variables. Probability parameters in the network are updated recursively. Root causes of occurring alarms are determined from the parent node set with the largest posterior conditional probability. The proposed method can remove negative effects of false and missing alarms in the nodes, handle the co-existence of multiple root causes, and detect the incompleteness of known root causes. Industrial examples are provided to illustrate the effectiveness of the proposed method.