Abstract

Large sized transformers are an important part of global power systems and industrial infrastructures. An unexpected failure of a power transformer can cause severe production damage and significant loss throughput the power grid. In order to prevent power facilities from malfunctions and breakdowns, the development of real-time monitoring and health prediction tools are of great interest to both researches and practitioners. An advanced monitoring tool performs real-time monitoring of key parameters to detect signals of potential failure through data mining techniques and prediction models. Asset managers use the result to develop a suitable maintenance and repair strategy for failure prevention. Principal component analysis (PCA) and back-propagation artificial neural network (BP-ANN) are the algorithms adopted in the research. This paper utilizes industrial power transformers’ historical data from Taiwan and Australia to train and test the failure prediction models and to verify the proposed methodology. First, PCA detects the conditions of transformers by identifying the state of dissolved gasses. Then, the BP-ANN health prediction model is trained using the key factor values. The integrated engineering asset management database includes nine gases in oil as input factors (N2, O2, CO2, CO, H2, CH4, C2H4, C2H6, and C2H2). After applying Principal Components Analysis, five factors from the Taiwan operational transformer data and six factors from the Australia data are identified. The integrated PCA and BP-ANN fault diagnosis system yields effective and accurate predictions when tested using Taiwan and Australia data. The accuracy rates are much higher (i.e., 92% and 96% respectively) when compared to previous result of 69% and 75%. This research is benchmarked against the DGA heuristic approaches including IEEE's Doernenburg and Rogers and IEC's Duval Triangle for the experimental fault diagnoses.

I represented my research-lab and presented at the International Asia Conference on Industrial Engineering & Management Innovation, 2013, National Taiwan University, Taiwan. 

Link to the conference website.

The conference presentation can be viewed here.

Milestones in this work

• Principal Component Analysis was applied for detection of condition of transformer by identifying the state of the dissolved gases and the key factor values.

• Industrial power transformers' historical data from Taiwan and Australia is utilized to train and test the failure prediction models and to verify the proposed methodology. 

• The Back Propogation - Artificial Neural Network health prediction model is trained using the key factor values and the diagnostic system yields effective and accurate predictions when tested using data from Taiwan and Australia.

References

• Avci, E. & Turkoglu, I. (2009) An intelligent diagnosis system based on principle component analysis and ANFIS for the heart valve diseases, Expert Systems with Applications, 36(2), 2873-2878.

• Booth, C. & McDonald, J.R. (1998) The use of artificial neural networks for condition monitoring of electrical power transformers, Neurocomputing, 23(1)-23(3), 97-109.

• Brusic, V., Rudy, G., Honeyman M., Hammer, J. & Harrison L., (1998) Prediction of MHC class II-binding peptides using an evolutionary algorithm and artificial neural network, Bioinformatics, Vol. 14, No. 2, 121-130.

• Ghunem, R.A., El-Hag, A.H. & Assaleh, K. (2010) Prediction of furan content in transformer oil using Artificial Neural Networks (ANN), IEEE International Symposium on Electrical Insulation (ISEI), San Diego, CA, USA, June 6-9, 1-4.