Power Transformer Fault Diagnosis And Initial Positioning Based On Vibrational Energy

According to analyzing the common fault type of power transformer and researching diagnostic methods for transformer winding and core fault,and their advantages and disadvantages.From finishing literature at home and abroad in recent years,it can be found that the partial failure of transformer fault diagnosis, so the necessity the latent fault online diagnosis is proposed.First consider the oil-immersed transformer as the study object,theoretical analysis on reasons of transformer vibration and the function between the internal fault with vibration are concluded based on detailed description of the structure of the composition of the oil-immersed power transformers,and the transmission route of the transformer vibration is also explored;the winding and core axial vibration model is established,from the view of theoretical analysis,it can be studied that the relationship among the transformer windings compaction degree, vibration acceleration amplitude and latent fault analysis respectively;a simple analysis of the impact on the tank surface vibration by the other components of the power such as transformer core,OLTC and cooling device.Then analyzing the characteristics of the power transformer vibration signals, a power transformer fault diagnosis method based on Hilbert Huang Transform is proposed.First, the vibration signal acquisition wavelet threshold de-noising, and then were on the winding, core fault signal are processed, and finally realize the transformer fault feature extraction. Experiments show non-stationary, time-varying signal effect becomes apparent with this signal processing method, can be a good reflection of the power transformer fault information, and can be used as the basis for power transformer fault diagnosis.Finally, the experimental method of vibration monitoring transformer fault is proposed, the position of sensor with a detailed description. A experimental program is designed,and the hardware system for diagnosis.