Fractal analysis on insulation degradation of polymer insulation material

The characteristics of polymer insulation materials often degrade remarkably due to surface discharge. In order to detect the degradation, a lot of methods have been taken to measure the currents and to deal with the data. However, the methods are not very suitable in some cases. In order to find a better way to assess the performance of the insulation materials, the fractal theory was adopted in this paper. According to the characteristics of statistical self-similarity and complexity of the discharge current, an approach was presented to find the fractal dimension based on the embedding space theory, state space reconstruction and G-P algorithm. The results show that the correlation dimension can identify the changes of the insulation characteristic, and the method is very useful at the diagnosis of the insulation degradation.