Machine learning for power transformer diagnostics

Abstract

Electric power system transformers are significant and high capital grid and network assets which require condition monitoring and pre-emptive assessment for ongoing resource management and regular maintenance. Among the myriad of methods suitable for transformer assessment, dissolved gas analysis (DGA) for oil-filled transformers is an effective and practical approach. Tapped oil sampling enables measurement of gas contamination which renders indices indicative of internal breakdown and faults which affect gas composition and internal oil dilution. Traditional methods for DGA assessment involve measurement of gas spectral contamination and human inspection. However, artificial intelligence (AI) methods and machine learning (ML) models have been proposed and widely used for fault diagnosis in recent years to eliminate and improve conventional DGA approaches.

Most existing AI diagnostic models have applied a supervised learning approach, based on adequate labelled field datasets. Due to the significant human intervention required, it is particularly challenging to collect adequate quantities of suitable DGA data, which contain a range and breadth of transformer conditions and can be easily labelled with known issues and emergent or established faults.

A new and hybrid method for transformer DGA condition monitoring using unlabelled datasets is proposed in this thesis. The work required construction of a large-scale fault database for the proposed ML method. Moreover, performance evaluation and assessment of common AI models, including artificial neural network (ANN), support vector machine (SVM), K-Nearest Neighbour (KNN) and Decision tree using the constructed database is presented.

Obvious differences in diagnostic assessment of transformer faults are a result of unequal quantities of datasets representative of the range of fault types. Fault classification using a large quantity of training data is better than comparable assessment using small quantities. In resolving problems with unbalanced datasets, this work investigated and proposed a classification method using a Gaussian process (GP). Comparative results between the proposed model and classical AI models are presented in this thesis to demonstrate effectiveness and improvements in diagnostic performance using GP multi-classification (GPMC), especially for small-quantity datasets. The reliability of the proposed ML method in comparison to conventional DGA approaches is also presented.

This thesis presents works investigating a hybrid AI model to resolve the difficulties with imbalanced datasets. Various data augmentation techniques are employed for data pre-processing, which balances the volume of training data in each fault. AI models then render diagnostic fault determination using processed datasets. An evaluation metric is used to analyse and compare the performance of different models. Experimental results demonstrate benefits for minority classes such as rare but significant transformer faults, including partial discharge (PD) and combination faults (DT), rendering improved diagnostic performance.

To forecast potential transformer faults, in addition to diagnosis at the instant a fault has occurred, a dynamic fault prediction model based on long short-term memory (LSTM) is proposed in this thesis. Time sequence information within DGA datasets is used to predict dissolved gas concentrations and the reliability of the proposed LSTM method is presented in comparison to common regression approaches in terms of error metrics. Additionally, different AI models are used to forecast future transformer health conditions using predicted results from LSTM outcomes, with comparative results presented.

Date of Award	Dec 2023
Original language	English
Awarding Institution	Queen's University Belfast
Sponsors	Engineering and Physical Sciences Research Council
Supervisor	Timothy Littler (Supervisor) & Xueqin Amy Liu (Supervisor)