High Impedance Fault Detection in Overhead Distribution Feeders Using a DSP-Based Feeder Terminal Unit

High impedance faults (HIFs) often draw too little current to activate conventional overcurrent protection, making the detection of a HIF event a difficult task. A practical and implementable solution to HIF detection and location is presented. In the proposed methodology, wavelet transform used for extracting features of HIFs signals and neural network used for HIF detection are used to develop a novel HIF detector called enhanced feeder terminal unit (FTU). The designed enhanced FTU is a high-performance HIF detector that is the retrofitting of the ready-made FTU with an embedded digital signal processor and a high-frequency current sensor. The enhanced FTU monitors the operation of distribution feeder system and sends a fault flag signal to feeder dispatch and control center when a HIF occurs. Based on the feeder automation mechanism, the HIF detection, isolation, and service restoration can be activated. Consequently, the power reliability can be significantly improved by reducing the outage area and repair time. The validation of the presented technique was evaluated through staged fault testing. Compared with the existing related research, the proposed enhanced FTU is available and feasible.