Journal of South China University of Technology (Natural Science Edition) ›› 2014, Vol. 42 ›› Issue (12): 112-119.doi: 10.3969/j.issn.1000-565X.2014.12.017

• Power & Electrical Engineering • Previous Articles     Next Articles

Improved Harmonic Current Detection for Three-Phase Four-Wire Active Power Filter

Wang Yu Xie  Yun-xiang   

  1. School of Electric Power,South China University of Technology,Guangzhou 510640,Guang dong,China
  • Received:2014-04-11 Revised:2014-08-23 Online:2014-12-25 Published:2014-11-17
  • Contact: 王 裕(1984-),男,博士生,主要从事有源电力滤波器研究. E-mail:xiaodou10@126.com
  • About author:王 裕(1984-),男,博士生,主要从事有源电力滤波器研究.
  • Supported by:

     广东省教育部产学研结合项目(2011B090400136)

Abstract:

Accurate and fast harmonic current detection plays an important role in determining the compensationperformance of active power filters.In this paper,an improved three-phase four-wire harmonic current detectionmethod is proposed by combining the traditional instantaneous reactive power theory with the FBD method.In thismethod,the algorithm complexity is reduced by cutting the unnecessary separation link of the zero-sequence cur-rent,and the matrix operations caused by the coordinate transformation in the traditional instantaneous reactivepower method are also reduced,thus improving the real-time and dynamic performance of the harmonic current de-tection.Moreover,the zero-axiscurrent is compensated separately and the balance control of upper and lower capac-itor voltages is conducted,so as to ensure the applicability of the active power filter of three-phase four-wire capaci-tors.In order to reduce the hysteresis error caused by the low-pass filter during the sampling and calculation and toimprove the harmonic detection accuracy,an improved moving average algorithm is adopted to replace the low-passfilter.Finally,the feasibility and superiority of the proposed method are verified through the simulation and experi-ment in a three-phase four-wire active power filter.

Key words: active power filter, instantaneous reactive power, FBD method, moving average algorithm

CLC Number: