Research On Separation Methods Of Background Harmonic In Power Supply System

The harm of harmonic is one of the main reasons to result in power quality problems, both power suppliers and users still have significant differcences on the reasons and responsibility of causing power quality down. Therefore, it is very important to separate background harmonic successfully from power supply system for both suppliers and users. It is also helpful to research single tuning filter in power system and to solve the problems of harmonic pollution in utility grid.In this paper, many separation methods of background harmonic in power supply system are described in detail above all, their advantages, disadvantages and principles are analyzed and summarized from the perspective of qualitative and quantitative respectively. The disadvantage of partial least-squares regression for assessing harmonic impedance and the background harmonic is indicated. Then, a new method based on the orthogonal signal correction and partial least-squares regression for assessing harmonic impedance and the background harmonic is given. New method combined with power system Thevenin equivalent and customer Norton equivalent, regression equations can be obtained from system parameter. Utilizing the signals of harmonic voltage and harmonic current measured synchronously at PCC, regression coefficients are worked out through the new method. Then the harmonic impedance and the background harmonic of system are got. The models of single-phase and three-phase power system are built by Matlab/Simulink respectively. Comparing the results with "Binary regression method" and "Partial least-squares regression method", it shows that the estimation error of the proposed method is lesser. Simulation analysis indicates that the proposed method is valid and feasible.