Intelligent Sensing Strategy for Grid Anomalies Using Power Line Communication Devices

The difficulty in reusing power line communication equipment for cable aging monitoring is the convergence of aging models and the presence of interference. This study introduces a novel intelligent sensing method that can simultaneously monitor the degree of water tree aging and cable grounding faults in distribution networks. An alternative approach uses backward derivation from channel estimation information from cable-dependent power line carrier devices. Further, it incorporates machine learning to sense dielectric constant anomalies to obtain the real-time status of cable anomalies. The simulation shows that the proposed algorithm can achieve 100% accurate monitoring in the case of pure aging in the grid; it can achieve more than 90% accurate monitoring and predict the aging depth under the coexistence of random load changes and fault disturbances.