| Utilities are already well aware of the impact of harmonic emission on power quality. Therefore, lots of algorithms for harmonic emission limits have been proposed to avoid exceeding the planning levels in the power system when all non-linear loads are working simultaneously. However, checking a posteriori whether the limits are fulfilled is a very difficult task. The problem is that most consumers are using linear-together with nonlinear loads. Harmonic currents flowing in the consumer connection are therefore resulting from harmonics originating in the general-and in the consumer system.The purpose of this article is dedicated to the problem of assessing the harmonic emission level from a distorting load. After analyzing the advantages and drawbacks of the reported algorithms, a novel approach is presented for evaluating the harmonic emission level of the customers system.Estimation of harmonic emission level depends on the accurate measurement of current and voltage at the common coupling point (PCC). Usually, the discrete Fourier transform (DFT), in the case of the fast Fourier transform (FFT), is the most commonly used method for electrical harmonic analysis. In addition, the DFT based method has been adopted by the IEC61000-4-7 standard. Nevertheless, the DFT is subject to errors from an effect known as spectral leakage and the fence effect. This effect occurs when the DFT is computed from a block of data that is not periodic. To improve the accuracy of harmonic parameters, weighting the sample data by the Hanning window is studied for better suppression of spectral leakage. And then, interpolated method is adopted to eliminate the errors caused by the fence effect. In practice, the Hanning window based interpolation method enhances the accuracy of power harmonic analysis greatly.For the multiple-linear problems of the independent variables in modeling, as well as the problems of resulting in wrong decisions since the coefficient information which should be retained is removed for mistake when the reported algorithms is applied for the estimation of harmonic emission level at PCC, the approach via principal component analysis is proposed in this article. Through the FFT processing the voltage and current sampling signals at PCC, the harmonic voltage and current which are as the dependent variable and the independent variable of the regression equation are obtained; then, an principal-component-analysis regression is utilized to solve the regression coefficient which is the system-side impedance; and then, the harmonic emission level of customer can be calculated according to its feature of harmonic current source. The principal component analysis based approach can extract the comprehensive variable which is the strongest explanatory for the independent variable and identify information and noise of system. Besides, the proposed method strengthens the correlation analysis of variables in statistical characteristics, overcoming the negative effects of variable correlation in system modeling greatly. Therefore, the harmonic emission level can be assessed precisely by the proposed method. The efficiency of the proposed method is verified by simulations and field test.Moreover, the customer side harmonic was researched in this paper in order to define the strengthened or weakened influence to the PCC harmonic. The criterion used to determine the strengthened or weakened effect of harmonic source is derived and simulation results show the validity of this criterion. |
PDF Full Text Request