Research On Traveling Wave Fault Phase Selection And Location For Common-tower Double-circuit Transmission Lines

In order to improve economic and social benefits and improve transmission capacity of transmission lines corridor,the common-tower double-circuit transmission lines have been developed rapidly.With a great number of types of faults and complex fault characteristics,mis-operation is easy to occur during the fault treatment of common-tower double-circuit transmission lines.In order to analyze fault information quickly and accurately,based on summarizing the traveling wave theory and phase mode transformation of transmission lines,and the characteristics of common-tower double-circuit transmission lines,the traveling wave fault phase selection and fault location methods are studied in the thesis.The derivation methods of different types of line phase mode transformation matrix of doublecircuit transmission lines are introduced.To solve the problem that single line mode component cannot reflect all fault types,a new phase mode transformation matrix which is suitable for commontower double-circuit transmission lines is proposed.The traveling wave process and the theory of refraction and reflection of traveling wave are introduced.The models of common-tower doublecircuit transmission lines are established using ATP-EMTP simulation software,and the characteristics of the voltage waveform with different circumstances are analyzed.The theory of wavelet teansform and empirical mode decomposition(EMD)are introduced.Based on EMD theory,in order to solve the problem of the mode mixing and ends effect,modified ensemble empirical mode decomposition(modified EEMD)is used to decompose signals.This method inherits the advantages of EMD and can quickly eliminate the influence of noise.The theories of self-correlation analysis method,cepstrum analysis,imformation entropy are introduced.The thesis introduces the concept of support vector machine(SVM).In order to achieve fast and accurate classification,the error-correcting output code classification machine is introduced.Compared with the traditional multi-classification method,the error-correcting output code classification machine has less computation and has the ability of error correction for support vector machine.In this thesis,wavelet transform and modified EEMD singular value decomposition and information entropy are combined into the study of traveling wave fault phase selection.The use of the fault voltage traveling wave for phase selection can quickly extract fault information.In this thesis,two kinds of signal decomposition methods are used to decompose the traveling wave voltage signal according to the frequency domain.The information entropy can reflect the distribution characteristics of the traveling wave energy between the fault phase and the non fault phase on different frequency components.On the basis of this,the fault phase selection is realized by using the error correcting output code classification machine.The simulation results show that the two phase selection methods can realize the fault phase selection quickly and accurately,and have strong antiinterference ability.Compared with the wavelet energy entropy fault phase selection method,modified EEMD singular value entropy phase selection method is more advantageous.In this thesis,the cepstrum analysis is introduced into the study of single terminal traveling wave fault location.The use of multiple signal classification(MUSIC)spectral estimation calculation is more complex,with low accuracy extraction of single terminal fault location method based on natural frequencies.The natural frequencies and wave head identification are combined,natural frequencies and wave head information are extracted by cepstrum analysis.Based on the line parameters of the simulation model,the frequency related wave velocity is obtained.On this basis,the single-end traveling wave fault location is realized.The simulation results show that the traveling wave fault location method using cepstrum analysis is simple,and has wide range of application,high precision and strong anti-interference ability.