论文信息 - Harmonic detection using feed forward and recurrent neural networks for active filters

Harmonic detection using feed forward and recurrent neural networks for active filters

Abstract In this study, the methods to apply the feed forward and Elman’s recurrent neural networks for harmonic detection process in active filter are described. Generally, Fourier transformation is used to analyze a distorted wave from power line, and a low pass filter is used to eliminate the fundamental wave before each harmonic component is detected. Due to this complicated process, the behaviour of active filter is delayed such that it is difficult to compensate harmonic in real time. In order to improve the processing speed and simplify harmonic detection process, the feed forward and Elman’s recurrent neural networks are used to detect harmonics from distorted wave instead of Fourier transformation and low-pass filter. We simulated the distorted wave including the 5th, 7th, 11th, 13th harmonics and used these neural networks to recognize each harmonic. The results show that these neural networks are applicable to detect each harmonic effectively.

[1] M. Rukonuzzaman. Adaptive Neural Network Based Harmonic Detection for Active Power Filter , 2003 .

[2] Jan Melkebeek,et al. Harmonic mitigation potential of shunt harmonic impedances , 2003 .

[3] Ward Jewell,et al. A new approach for suppressing harmonic disturbances in distribution system based on regression analysis , 2001 .

[4] Mototaka Sone,et al. An application of neural network for harmonic detection in active filter , 1994, Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).

[5] Simon Haykin,et al. Neural Networks: A Comprehensive Foundation , 1998 .

[6] W. E. Reid. Power quality issues-standards and guidelines , 1994 .

[7] Virginia L. Stonick,et al. A DSP controlled resonant active filter for power conditioning in three-phase industrial power systems , 2002, Signal Process..