Function analysis based rule extraction from artificial neural networks for transformer incipient fault diagnosis

• Published in: International journal of electrical power & energy systems 2012-12, Vol.43 (1), p.1196-1203
• Main Authors: Bhalla, Deepika, Bansal, Raj Kumar, Gupta, Hari Om
• Format: Article
• Language: English
• Subjects: Applied sciences; Approximation; Artificial neural networks; Electrical engineering. Electrical power engineering; Exact sciences and technology; Extraction; Fault diagnosis; Knowledge rule extraction; Learning theory; Linear equations; Neural networks; Testing. Reliability. Quality control; Transformer fault diagnosis; Transformers; Transformers and inductors
• ISSN: 0142-0615; 1879-3517
• DOI: 10.1016/j.ijepes.2012.06.042

► ANN is trained using data of gases dissolved in the transformer oil as input. ► Hyperbolic tangent function of the NN is approximated by three piece linear approximation. ► The input space is split into three subregions and there is a linear equation for each subregion. ► Equations are used to predict the fault type in a transformer. ► Promising results were obtained when applied to case studies for incipient fault diagnosis in an transformer. Dissolved gas analysis (DGA) has been widely used for fault diagnosis in a transformer. Artificial neural networks (ANN) have high accuracy but are regarded as black boxes that are difficult to interpret. For many problems it is desired to extract knowledge from trained ANN so that the user can gain a better understanding of the solution arrived by the NN. This paper applies a pedagogical approach for rule extraction from function approximating ANN with application to incipient fault diagnosis using the concentrations of the dissolved gases within the transformer oil, as the inputs. The proposed method derives linear equations by approximation the hidden unit activation function and splitting the input space into subregion. For each subregion there is a linear equation. The experiments on real data indicate that the approach used can extract simple and useful rules. Transformer incipient fault diagnosis can be made that matches the actual fault present and at times the predictions better than those of the IEC/IEEE method. The rule sets generated have been successfully checked for accuracy of predictions by applying them to case studies.