Smart Monitoring of Partial Discharges in Power Apparatus Using UHF Sensor

Partial Discharge (PD) activity is a pre-cursor for insulation degradation which may eventually lead to catastrophic failure of power apparatus with severe social and economic consequences. Electromagnetic emissions gathered over ultra-high frequency (UHF) range (0.5 to 3 GHz) during PD events is widely investigated for health monitoring of high voltage (HV) apparatus. However, large bandwidth of PD induced UHF signals requires high speed digitizer and large storage space which have restricted field deployment of UHF technique for online PD detection. In this work, we present UHF based PD pulse detection framework for remote monitoring and logging of faults in dielectric insulation using ThingSpeakTMcloud. PD events captured by a calibrated UHF sensor from HV PD test cell using the proposed online PD detection framework indicated throughput as high as 83.8% compared with reference UHF sensor measurements gathered by a digital storage oscilloscope. PD detection throughput was observed to be limited by the envelope detection threshold.Vin(10 mV) and cumulative propagation delaytpd(∼1μs) in the architecture implementation. The proposed UHF sensing technique is low cost compared to conventional approaches, detects a wide range of insulation defects, and scalable for smart monitoring of HV assets using Internet of Things (IoT).