This study proposes a complementary fault identification scheme for triple fed high-voltage transmission lines (TFHVTLs). The current signals are captured from a single end measuring system of the TFHVTL. Then, the captured current signals are decomposed using wavelet multi-resolution analysis technique. The proposed scheme is based on two criteria. The first criterion is dependent on the level of detail coefficients while the second one is dependent on the deviation of detail coefficients. The decision is obtained using an OR-logic of the two criteria. Also, this study introduces the stability and sensitivity concepts for the fault identification process. In addition, a suggested fault identification sensitivity index is proposed to measure the efficacy of the proposed scheme. Also, different types of mother wavelets are assessed to show that the best mother wavelet, Haar, has high fault detection and classification performance. An ATP/EMTP package is used to simulate the tested part of the 500 kV Egyptian grid. The proposed scheme performance is corroborated under different operating conditions for load changes, fault locations, sampling frequencies of the monitored current, fault resistances, fault types, and inception angles. The proposed scheme is tested for series compensated TFHVTL at different compensation levels. |
