Data mining for oil‐insulated power transformers: an advanced literature survey

Abstract

Knowledge discovery in database and data mining (DM) have emerged as high profile, rapidly evolving, urgently needed, and highly practical approaches to use dissolved gas analysis (DGA) data to monitor conditions and faults in oil‐immersed power transformers. This study reviews different DM approaches to oil‐immersed power transformer maintenance by discussing historical developments and presenting state‐of‐the‐art DM methods. Relevant publications covering a broad range of artificial intelligence methods are reviewed. Current approaches to the latter method are discussed in the field of DM for oil‐immersed power transformers. In this paper, various DM approaches are discussed, including expert systems, fuzzy logic, neural networks, classification and decision, and hybrid intelligent‐based diagnostic systems that apply the DGA database. © 2012 Wiley Periodicals, Inc. This article is categorized under: Application Areas > Industry Specific Applications

A Multinomial DGA Classifier for Incipient Fault Detection in Oil-Impregnated Power Transformers

“…Fault gases can be categorized into three [7,8,[33][34][35][36][37][38] The quantity and ratios of these gases will vary depending on whether a fault is developing in the transformer or not. A critical evaluation of the variation of gases present can precisely determine the underlying condition and the extent to which internal damages have been occasioned [39]. The leading causes of gas formation within an inservice transformer are either of thermal or electrical origins.…”

Power Transformer Failure Prediction: Classification in Imbalanced Time Series

“…In this section some literature on the topic of using data mining (DM) techniques in power transformer data and fault prediction is exposed. Huang, Huang, and Sun (2012) reviewed different DM approaches to oil-immersed power transformer maintenance. It focused more on diagnosis, identifying the cause of failure after the failure has occurred, with brief mentions to prognosis of faults.…”

Factorial analysis for ageing assessment of in‐service transformers

“…The techniques include graphical analyses like scattered plot and curve fitting, as well as statistical analyses such as hypothesis tests and correlation coefficients [3][4][5][6][7][8]. Apart from that, mathematical techniques such as weightings-based health index formula, fuzzy logic, artificial neural network, principal component analysis, analytic hierarchy process, grey relational analysis, support vector machine and particle swarm optimisation have also been proposed for condition assessment of in-service power transformers [4,[9][10][11][12][13][14][15].…”