Analysis of Synchronous Generator Internal Faults Based on Fractal

The parameter waveforms of synchronous generator are related closely to their operating conditions, which can reflect and foresee some certain states or faults to occur in generator. The chaos and fractal theory is used to synthetically analyze parameter waveforms in this paper. The steady and transient process of synchronous generator with internal fault has been investigated based on the rescaled range analysis. Through fractal dimension analysis of the simulative and experimental data respectively, two main useful laws are obtained. Firstly, transient waveform´s fractal dimension is a quantitative representation of generator´s internal fault type and have great relevancy with its spectrum distribution. Secondly, the steady and transient state fault pattern can be detected by using R/S analysis method to some extent, but it should combine with other method before it can be applied into practice. These results provide new criterions for the protection and fault detection of the generator with internal fault