Learning Models of Plant Behavior for Anomaly Detection and Condition Monitoring

Abstract

This version is available at https://strathprints.strath.ac.uk/17089/ Strathprints is designed to allow users to access the research output of the University of Strathclyde. Unless otherwise explicitly stated on the manuscript, Copyright © and Moral Rights for the papers on this site are retained by the individual authors and/or other copyright owners. Please check the manuscript for details of any other licences that may have been applied. You may not engage in further distribution of the material for any profitmaking activities or any commercial gain. You may freely distribute both the url (https://strathprints.strath.ac.uk/) and the content of this paper for research or private study, educational, or not-for-profit purposes without prior permission or charge.Any correspondence concerning this service should be sent to the Strathprints administrator: strathprints@strath.ac.ukThe Strathprints institutional repository (https://strathprints.strath.ac.uk) is a digital archive of University of Strathclyde research outputs. It has been developed to disseminate open access research outputs, expose data about those outputs, and enable the management and persistent access to Strathclyde's intellectual output. Abstract-Providing engineers and asset managers with a tool which can diagnose faults within transformers can greatly assist decision making on such issues as maintenance, performance and safety. However, the onus has always been on personnel to accurately decide how serious a problem is and how urgently maintenance is required. In dealing with the large volumes of data involved, it is possible that faults may not be noticed until serious damage has occurred. This paper proposes the integration of a newly developed anomaly detection technique with an existing diagnosis system. By learning a Hidden Markov Model of healthy transformer behavior, unexpected operation, such as when a fault develops, can be flagged for attention. Faults can then be diagnosed using the existing system and maintenance scheduled as required, all at a much earlier stage than would previously have been possible.

The role of circumstance monitoring on the diagnostic interpretation of condition monitoring data

“…Brown et al suggest six phase segments [12]; three per half cycle corresponding to the rising positive segment, the positive peak, the falling positive segment, and falling, peak, and rising segments of the negative half cycle. For each of these phase segments, statistical features of mean, standard deviation, and kurtosis are calculated.…”

Development of an intelligent system for detection of exhaust gas temperature anomalies in gas turbines

“…For this reason, previous work has combined these approaches (5) . In the cited case, an anomaly detection algorithm is used first to detect whether a fault has occurred and Development of an intelligent system for detection of exhaust gas temperature anomalies in gas turbines A D Kenyon, V M Catterson and S D J McArthur if so, delegate the task of identifying the particular fault to a fault diagnosis algorithm.…”

On-line Transformer Condition Monitoring through Diagnostics and Anomaly Detection

“…These include data-driven techniques [11], knowledge-based interpretation [12], and anomaly detection [3]. It was discovered that whilst, individually, the data-driven diagnostic techniques could each diagnose faults in the transformer to some degree, certain diagnostic approaches would lead to a more accurate diagnosis of particular types of fault.…”

“…To further improve the original COMMAS, an anomaly detection agent was created as the first stage of PD analysis [3]. This agent learns normal PD behavior for individual transformers, after which the detection of anomalous events can warrant classification by the diagnostic agents and explanation through the knowledge-based approach.…”

“…Since each transformer behaves individually, machine learning techniques can be applied to learn what normal behavior is for a particular item of plant [3]. If the item's behavior diverges from normal, it can be identified that an anomalous event has occurred within this plant item.…”