Adaptive Partial Discharge monitoring system for future smart grids

This paper proposes an asset monitoring technique for smart grid applications. An adaptive wavelet-based technique was utilized to efficiently handle large volumes of data and built a relationship between the monitored data and the equipment electrical insulation level. The proposed technique has the ability to detect the early stages of Partial Discharge (PD) activities embedded in high noise and extract pulse shape PDs using a manageable data set that can be handled by Wireless Sensor Networks (WSNs). The proposed technique was investigated using laboratory and field data sets that captured by non-intrusive high-frequency currents transformers and acoustic emission (AE) sensors to accommodate aging and operation stress and gave promising results in monitoring the early stages of PD activities.