Fault Detection And Classification In Transmission Lines Based On Empirical Wavelet Transform And Learning Vector Quantization Neural Network

With the continuous expansion of UHV power grid construction,the power system in China has been a long-distance,large-capacity,low-loss electricity transmission model,which leads huge economic benefits to social production and livelihood.However,the increase of transmission voltage grade and complicated power grid structure impose higher requirements on stability and security of modern power systems.Transmission lines are the carrier as well as an important part of the power system.If a short-circuit fault occurs,it will bring huge losses to the national economy and people's livelihood.Therefore,establishing an effective transmission line fault detection and classification model has important practical significance for ensuring the stable operation of the power system.In this paper,the transient currents of transmission lines are taken as the research object.According to signal processing technology and artificial intelligence algorithm,a fault classification method for transmission lines based on empirical wavelet transform and learning vector quantization neural network is proposed.The specific research content is as follows:An approach to fault feature extraction is presented,which based on Empirical Wavelet Transform.Compared with traditional signal processing techniques,empirical wavelet transform can adaptively decompose signals.Simulation analysis also shows that empirical wavelet transform can effectively suppress the modal aliasing phenomenon of the empirical mode decomposition.In addition,it can accurately analyze the transient current fault information.Based on empirical wavelet transform theory,three kinds of fault feature extraction methods are studied and proposed.Fault information is effectively and accurately analyzed and normalized from the frequency domain and time-frequency domain of the signals.In the process,the 32-dimensional fault feature vector is finally obtained,which truly reflects the fault current characteristics of the transient current.Finally,an optimized learning vector quantization neural network based on artificial bee colony algorithm is proposed and used as a transmission line fault classifier.With the introduction of artificial bee colony algorithm,learning vector quantization neural network solves the problem of sensitivity to initial weight value and improves the classification performance.A 500 kV EHV transmission line simulation model is made by using Matlab,which produces a large amount of data to train the neural network classifier.And next use the trained fault classification model to classify faults.The final simulation results show the effectiveness of the fault classification model of the transmission line.