LECTURE NOTES IN COMPUTER SCIENCE

Abstract

In this study, the method to apply the Elman's recurrent neural networks for harmonic detection process in active filter is proposed. The feed forward neural networks were also used for comparison. We simulated the distorted wave including 5(th), 7(th), 11(th), 13(th) harmonics and used them for training of the neural networks. The distorted wave including up to 25(th) harmonics were prepared for testing of the neural networks. Elman's recurrent and feed forward neural networks were used to recognize each harmonic. The results show that these neural networks are applicable to detect each harmonic effectively.