A Diagnostic Approach to Power Transformers Based on Genetic Wavelet Networks Sample

For the purpose of fault diagnosis of power transformers, an approach using genetic wavelet networks (GWNs) is proposed in this paper. The GWNs have a three-layer structure which contains wavelet, weighting and summing layers. By genetic algorithm (GA), the GWNs tune the network parameters, translation and dilation in the wavelet nodes and the weighting values in the weighting nodes automatically. With the global search abilities of the GA and the multi-resolution features of the wavelet, the GWNs can identify the complicated relations of dissolved gas contents in the transformer oil to corresponding fault types. As revealed in the experimental results, the proposed approach outperforms conventional methods in both diagnostic accuracy and construction time.