Power Transformer Transient Based Protection Using Wavelet De-Noising

A novel and effective technique for discriminating fault current from inrush current based on wavelet de-noising is presented in the paper. Internal fault current, external fault current and inrush current exhibit distinct characteristics through the detecting of fault skips. Based on these, a criterion which proved to provide to be reliable, fast, and computationally efficient has been developed. The dynamic simulation data were used to test the performance of the technique as to its speed of response, computational burden and reliability. The algorithm developed is still effective when CT saturation is presented, thereby, it can be used as an attractive and effective approach for alternative protection technique for large power transformers.