부산대학교 도서관

Aerial Bundle Cables (ABCs) have been used for overhead power distribution in many parts of the world. ABCs are of great interest to metropolitan areas with electrical pilferage problems, as their inherent insulation offers better protection. One such instance is M/s KE deploying ABCs in the coastal city of Karachi. The ABCs experienced rapid degradation due to the moist, humid, and harsh environment of Karachi. However, the degradation is not visually observable due to insulation, and in turn, compromises early detection. Non-Destructive Testing (NDT) techniques offer a solution for condition monitoring of the ABCs for degradation detection. A historical database of appropriate NDT and environmental data enables degradation trend prediction. This research work reports a degradation trend prediction scheme for ABCs using Recursive Neural Network (RNN) based Artificial Intelligence (AI) model. The reported work uses custom-acquired real NDT data of in-service ABCs. Promising results show the efficacy of the AI-based prediction scheme.