High impedance faults (HIFs) are difficult to be detected through conventional protection relays such as distance relays. When a conductor makes a contact with a poor conductive surface, the
resulting level of fault current is usually lower than the nominal current of the system at that fault location. Therefore, conventional relay system will not be able to detect and trip HIFs. The failure of HIF detection may lead to potential hazards to human beings and
fires [1]. HIFs on electrical transmission and distribution networks involve arcing and/or nonlinear characteristics of fault impedance, which cause cyclical pattern and distortion. Therefore, the objective of most detection schemes is to evaluate the special features
in patterns of the voltages and currents in HIFs. Several researchers in recent years have presented many techniques aimed for detecting HIF more effectively. These techniques
include discrete wavelet transform with other different methods [1–3], down-conductor fault detection and location via a voltage based method [4], and development of a fuzzy inference system based on genetic algorithm [5]. This paper describes a fault detection technique for HIF in transmission line. This technique involves capturing voltage signals generated in transmission lines under different HIFs via coupling capacitive voltage transformer. The detection process is performed through signal decomposition, thresholding the absolute sum values of the wavelet transform coefficients for one cycle in a moving window scheme against a real time estimated threshold value.