Integration of online proactive diagnostic scheme for partial discharge in distribution networks

Partial discharge (PD) diagnostic is considered as the main concern while making condition assessment plan for medium voltage assets. PD detection and localization in multi-section (straight and branched power lines) MV network is difficult by using the conventional Time Domain Reflectometry and Time Difference of Arrival methods. It is due to interconnected cables of different lengths and properties which make the interpretation of PD signals quite complex as compared to single cable routes. This paper presents an online technique of PD localization in MV cable network. PD activity emits current pulses propagating away from the PD site. Polarity of the detected PD pulses with reference to polarity of the supply voltage determines the direction of arrival of these pulses. Comparison of the polarity of PD pulses identifies the faulty section of the overhead covered conductor line as well as the cable network. A detailed discussion has been presented to address the practical issues of sensor implementation and polarity comparison in order to ensure the accuracy of polarity based direction of arrival (DOA) assessment during field application. Evaluation of DOA is based on directionally calibrated sensors. Allocated induction sensors are employed along the feeder for integrated implementation of proposed technique in accordance with the distributed agent for improving the network reliability.