Power Line Transmission Fault Modeling and Dataset Generation For AI Based Automatic Detection

Fault in power transmission line need to be immediately solved to avoid power outage. Artificial intelligence (AI) based automatic fault detection emerges as simple and accurate solution to anticipate this failure. The main problem in AI application is the requirement of large amount of training data. However, acquiring real faults data are costly and technically difficult. Furthermore, distinguishing the type of fault is a complicated due to high data processing requirements. In order to solve above problems, this paper presents a method to generate fault dataset for AI training. Proposed method models all types of fault. The proposed fault model is implemented in Simulink. In order to generated dataset, we input various condition of fault to the model. The generated dataset then evaluated using various kind of AI methods. The experimental result showed high accuracy AI can be trained using proposed generated dataset.