Research On The Harmonic Detection Based On Wavelet Transform And Modern Spectral Estimation

Along with the social modernization development, power system throughout all walks of life, plus a lot of nonlinear components in use, harmonic problems become more and more complex. Power system fault by previous experience can be seen, in part by caused by harmonics. Therefore, in order to improve and enhance the quality of electric energy, must take effective technical means. To be able to accurately detect the fault location and fault cause. In view of such problems, the existing research results foundation, on the power system harmonic detection and control method for further study.Fourier transform was first applied to harmonic detection is the most classic. But with the development of power system is becoming more and more complex, add random harmful noise seriously affects the Fu Liye transform detection effect, and Fu Liye transform method can not take into account the frequency domain and time domain. Plus the amount of calculation is relatively large, which leads to it can’t meet the need of real-time detection. In view of this situation, choice of wavelet transform harmonic detection method. The simulation results prove that, for the vast majority of the harmonic can have good detection effect. Especially the significant role in the elimination of noise.Because wavelet transform has good localization characteristics of time domain and frequency domain, it can in the modern harmonic detection is widely used in the cause of. Modern spectral estimation algorithm of Prony in theory but also has higher frequency resolution, to isolate the various frequency components. By using the method of harmonic detection and through the Matlab simulation can be found, in no noise case, Prony algorithm has good detection effect. But in the adding noise simulation can be found under the influence of the noise is very big. In the classical Prony algorithm based on, in view of the actual environment of the traditional Prony algorithm, this method not only reduces the amount of computation, but also in low noise environment without de-noising can accurate analysis.According to the result and the previously mentioned wavelet transform in signal denoising is unique, therefore, comprehensively by using wavelet transform and modern spectrum estimation by combining the harmonic detection method. Through making up for the shortcomings of both, with the advantages of both. Through repeated simulation, draw the idea is to meet the requirements of the. And a good detection effect.This article is in view of the above described the thought is validated by simulation. The simulation results show that this kind of thought and feasible. The harmonic detection becomes more simple, efficient, fast.