Skip to Main Content
The empirical mode decomposition method can separate the transformer winding vibration signal into finite modes, then disclose the running status of transformer, whereas, the decomposing results of winding vibration are disturbed by the end distorting. To optimize the empirical mode decomposition, the time series modeling and predicting were introduced to extend the signal, and non-uniform B-spline curve was provided for overshoots and undershoots of polynomial spline at the same time. Associating with predicting vibration data beyond observation series by model, envelope fitting of data inside and outside observation series based on non-uniform B-spline curve is got. Envelope fitting based on predicting model and B-spline interpolation curves alleviates large swings. Left by themselves, the end swings can eventually propagate inward and corrupt the whole data span. The decomposing results of transformer winding vibration show that intrinsic mode functions. It is demonstrated by actual analysis that the modified empirical mode decomposition method is one effective way.