IoT Application in Transformer Fault Prognosis Using Vibration Signal

摘要

Among various methods developed to monitor transformer condition and diagnose transformer electrical and mechanical issues, vibration monitoring is economical and suitable for real-time implementation. Amplitude, intensity, harmonic components and direction of the vibration signals can assist in identifying the condition of the transformer windings. In this study, a model capable of predicting the transformer short circuit in initial stage is developed using vibration signals. Moreover, insulation damage, and transformer over and under excitation is also studied. Real-time diagnosis is achieved using cloud computing and IoT protocols which enable data transfer, storage, and retrieval for all authorized users.