Study About Real-time Of Apf Harmonic Detection

In recent years, with the wide application of power electronics, electric energyhas been more fully utilized. However, power electronic devices with nonlinearcharacteristics of their own will make the grid voltage and current distortion. Thesehighly nonlinear devices increasing in volume and capacity have made power systemsharmonic pollution get worse. They have become the pollution source of affecting thepower quality and caused great impact on power safety and economic performance ofthe power system. At the same time electric equipment for power system and userterminal has increasingly demand of power quality. So the solution to harmonicpollution in power system control has important practical significance.The relevant basic theory of Active power filter, as an effective method forharmonic pollution in power system, is relatively mature. But there still are manyflaws, for example, how to improve the accuracy of harmonic detection and real time,how to make the compensation control algorithm more flexible, etc. There are manyresearchers doing a lot of research on these areas.This article focuses on the harmonic detection of active power filter. First, someof the main harmonic detection algorithms are introduced, including the Fast FourierTransform method, instantaneous reactive power method and harmonic detectionbased on neural network, etc. There is a brief introduction to principles andcharacteristics of these algorithms. Next the factors resulting on the delay of activepower filter are analyzed and there is qualitative and quantitative analysis of theimpact of delay on the harmonic detection. On the basis of the first two parts,combined with the traditional harmonic detection algorithm, some improvements havebeen made to enhance the real-time and accuracy of harmonic detection. Based ontheoretical analysis, experiments of improved algorithms are made through softwareand hardware. They further verify the impact of delay on harmonic detection and theenhancement of real-time and accuracy of harmonic detection resulting from theseimproved algorithms. The experimental results demonstrate their good performance.