Аннотация
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.