Harmonic detect based on BP network and Fourier progression
A harmonic detect method based on BP network is introduced. BP output in Fourier format equals with Fourier progression, so BP network is designed to be a SISI system. Hidden layer is composed with Fourier trigonometric function as activation function. Gradient descent regulates weight to reduce error. To insure the convergence of the BP network, learning step η is settled according the Lyapunov theorem. Simulation result shows that it can acquire harmonics magnitude and phrase precisely and quickly.
Harmonic BP network Fourier Gradient descent
DONG Tao CHENG Pei-yuan FAN Bo WANG Dong-xu
The Missile Institute, Air Force Engineering University, Sanyuan, Shaan xi, China, 713800
国际会议
第八届国际测试技术研讨会(8th International Symposium on Test and Measurement)
重庆
英文
1512-1515
2009-08-01（万方平台首次上网日期，不代表论文的发表时间）