Research Of Power Disturbance Signal Denoising And Detection Based On Wavelet

With the continuous improvement of the economic level in today's society,energy resources has become a widely used resource, application of energy resources can reflect a country's level of development. In recent years, a large number of non-linear equipments, as well as with impact and imbalance of the load into the power grid, will cause serious influence on network performance.The modern industrial production of power quality requirements continue to improve, so that people gradually deepen the awareness of the importance of power quality. In order to effectively solve the problem of power quality,improve the power quality of the present situation, we must first have problems related to the analysis and study of power quality.This paper introduces the power quality research background, significance,basic concept and current status of research. Mainly for transient power disturbance signal to noise and disturbance localization and denoising timeconsuming problems do some related research work, put forward a kind of based on dual tree complex wavelet transform and maximum after posterior estimate method of combining the power disturbance signal to noise and enhance the small wave to disturbance location positioning, and achieved good results.Specific methods are as follows: first by the correlation processing method of noisy disturbance signal of rough denoising, and dual tree complex wavelet transform decomposition and reconstruction of power disturbance signals. Due to the strict sampling characteristics of wavelet transform. In the application will bring a lot of problems and dual tree complex wavelet transform is half the sampling characteristics of dual tree structure can well make up for the lack of real wavelets. The experimental results show that after the relevance of pretreatment, after the maximum posterior estimate method of dual tree complex wavelet decomposition coefficient threshold processing, reconstructioncoefficients obtained after the signal to noise ratio is relatively high, mutation information retention ability is good.The wavelet transform of ascension can obtain the advantages of the same address calculation, in computer operation can save memory in a large extent overhead and now commonly used wavelet filter decomposition processing, can be a substantial increase in efficiency of the wavelet transform. According to the wavelet theory, if the increase in the length of the filter, the operation speed can be doubled compared to the conventional wavelet. In this paper, after denoising the disturbance signal of power, take the 9/7 lifting wavelet lifting scheme to detect the disturbance to the starting and ending time. Simulation experiments show that the wavelet method is simple, fast, accurate positioning, low timeconsuming, overcomes the drawbacks of the traditional wavelet transform,proves the superiority of lifting wavelet.