Research On Detection Method Of Power Grid Distortion Signal Based On Wavelet Transform And Curve Fitting

With the increase of non-linear load in power grid,the signal of power grid is superimposed on the noise signal to form the distorted signal of power network.The detection accuracy of the noise signal directly determines the accuracy of the electric energy measurement in the power grid distorted signal.Power grid noise signal has the characteristics of time varying and frequency complex.It reduces the accuracy of the electric energy measurement.The interest of relevant departments of national electrical power has been loss.Most of the existing research on distorted signal is based on the energy distribution of the noise frequency.Selecting the appropriate layer of wavelet transform method to decompose and reconstruct,it leads to a high accuracy of noise detection.Incorrect decomposition level can reduce the accuracy of the noise detection in the distorted signal of the power grid.To solve these problems,it needs to propose a new method to analyze and detect the noise signal of power network.This paper focuses on the accuracy of distorted signals steady and non-steady state noise detection.In order to improve the accuracy of the noise signal detection,this paper proposes a detection method of the noise signal based on wavelet transform and curve fitting.First of all,the power grid distortion signal is decomposed and reconstructed by the wavelet transform,obtaining the fundamental signal.Then the reconstructed fundamental signal curve fitting by Fourier series.Finally,the non-stationary distorted signal minus sine signal is the non-stationary distortion noise signal.In order to verify this paper's method.This paper provides the steady state power grid distortion signal model and the nonsteady state power grid distortion signal model and the specific research contents areas follows:1.Study the mathematical model of noise signal distortion under the condition of power grid,through the establishment of mathematical model of steady and nonsteady grid distortion.Analyzing limitations of the noise effects on the power grid quality and the existing noise detection method.2.Searching for suitable noise detection method on power grid distortion signal for unknown noise and energy distribution.First the wavelet transform is applied to grid distortion signal for decomposition and reconstruction,then using the line fitting method and the normalized reconstruction to the fundamental signal synthesizing standard fundamental signal.Obtain the noise signal with high accuracy.3.The typical power grid distortion signal superimposed several different types of noise.Simulating above signals through this method,compare the results with single wavelet transform.Analysis the advantage of this method compared with the traditional method.The experimental results are the correlation coefficient between the reconstructed signal and the original signal is more than 0.95,which means the consistency between the two signals is above 95% in the field of distorted signal analysis and detection.Under the same decomposition level this method can improve the consistency of the original signal reconstructed by the wavelet transform method.And between different decomposition layers,the detection result is not changed.It has high signal detection stability of the power grid distortion signal.