Study On The Mechanism And Identification Method Of Complex Sympathetic Inrush In Transformer

Power transformer is an important equipment in each link of power system,which plays a key role in the stability and continuity of power supply.In the research of power system transformer relay protection,the misoperation of differential protection caused by magnetizing inrush current has always been a research hotspot,but the sympathetic inrush current has not been paid enough attention because of its concealment.In recent years,there have been many accidents of no-load closing transformers causing the appearance of adjacent transformers sympathetic inrush current,which caused the main protection to malfunction.In some cases,it also causes the superior line backup protection and overcurrent protection to misoperate.In this context,sympathetic inrush current has been paid more attention and studied by more scholars and researchers.Transformer protection is still lack of identification methods for sympathetic inrush current,so most of the protection site is still using identification methods for excitation inrush current.As a result,the protection cannot correctly identify and act when sympathetic inrush current occurs,which brings hidden dangers to the safe operation of the entire power grid.Therefore,the research on the sympathetic inrush current caused by no-load transformer closing and the identification method are put forward to improve the safe and stable operation of the transformer.In this context,this paper studies the mechanism of complex sympathetic inrush current generation,establishes the corresponding simulation model.In this paper,a method to identify sympathetic inrush current,magnetizing inrush and internal fault current is proposed,which can identify them simultaneously.The main research work and conclusions of this paper are as follows:(1)The model of transformer complex sympathetic inrush current which is more suitable for actual operation is studied.The time-domain expression of the flux linkage is solved by using Laplace and inverse Laplace transforms.The mechanism of complex sympathetic inrush current generation is analyzed more accurately,which provides a theoretical basis for modeling in the following chapters.(2)The complex sympathetic inrush current models were built by the electromagnetic transient simulation software,and the internal fault current model was built.The data of sympathetic inrush current,magnetizing inrush and various fault current are obtained by simulation.Considering the influence of different factors on sympathetic inrush current,the inrush current samples identified by this method are more comprehensive.In addition,the engineering case of protection misoperation caused by sympathetic inrush current is analyzed and studied.(3)In this paper,a new method of sympathetic inrush current identification based on wavelet packet and WOA-KELM model is proposed.Wavelet packets are used to decomposed and reconstruct current signals,the sympathetic inrush current,magnetizing inrush current and internal fault current are decomposed at different scales and focused at various details.The eigenvector is composed by calculating the relative energy ratio.The machine learning KELM model was selected to map the input data to a higher dimensional space for classification.At the same time,the whale population optimization algorithm WOA was introduced to determine the optimal penalty factor c and kernel function parameters σ in KELM.The results show that the overall recognition rate of WOA-KELM model in the identification and sympathetic inrush current,magnetizing inrush current and internal fault current reaches 99.44%.Compared with other models,the recognition rate is higher and the calculation speed is faster.