Power Quality Disturbance Identification Based On Improved S Transform And Support Vector Machine

With the continuous progress of science and technology,a large number of nonlinear loads are connected into the power system,which result in a large number of complex power quality problems,so the power quality disturbance classification and identification is becoming the focus of attention.Identifing all kinds of single and compound power quality disturbance is a prerequisite to solve th e power system fault,harmonic shared responsibility and power based quality assessment of follow-up study,which has important scientific significance.Firstly,this paper introduces the basic index of power quality,modeling and simulation of the voltage sags,harmonics and other common disturbance,using S transform in power quality disturbance detection method based on multi resolution analysis,the characteristics of the signal time-frequency analysis,get the fundamental amplitude and frequency of maximum amplitude,the signal curve description check the time and frequency characteristics of harmonic?Secondly,this paper improves the S transform window function,improved S transform into amplitude and index adjustment coefficient,accordin g to the characteristics of different frequency disturbance,considering the influence factors of noise,puts forward the time-domain waveform described four waveform index,rough set theory and the index weight of each wave of subjective and objective weighting method in order to find the optimal combination of the optimal frequency waveform adjustment coefficient,lay the foundation for the classification and identification of disturbance.Finally,accurate extraction of characteristics of the optimal wav eform,the global and local kernel function with mixed kernel function of support vector machine classification,by changing the mixed kernel function adjustment coefficient can maximize the classification accuracy for different types of data,effectively improve the generalization ability and classification accuracy of classifier.Based on the Lab VIEW environment,prepared power quality disturbance classification software,can directly to the power system disturbance signal detection and analysis,then the actual substation each branch voltage signal are analyzed according to the actual situation of engineering,proving the feasibility of the proposed method.