Intelligent Substation Transformer Fault Diagnosis Method Based on Voiceprint Signal
The technology of intelligent substations is flourishing,and the safe operation of transformers is a necessary prerequisite to ensure the operation of substations.However,due to the harsh working environment of power transformers,the collected data often has problems such as noise and interference,which puts forward strict requirements for the fault diagnosis technology of intelligent substation transformer equipment.In addition,some nonlinear faults in transformers are difficult to detect,and traditional diagnostic methods cannot detect them.In order to solve the problem of transformer fault diagnosis in intelligent substations,a transformer fault diagnosis method based on voiceprint signal is proposed.A broadband background noise suppression method is based on the fusion of wavelet transform and peak extraction.On the basis of fully suppressing broadband background noise,sound signals are extracted,and a transformer fault diagnosis model based on Mel spectrogram is constructed.Finally,intelligent perception and effective recognition of transformer faults in intelligent substations are achieved.
transformerintelligent substationfault diagnosisvibration signal
NETL
NSTL
万方数据
