Identifying and controlling deadband systems using neural networks
暂无分享,去创建一个
We present a solution to identify and control a system containing a deadband feature. After describing the deadband, we show how to compute a convenient architecture for the neural network in charge of the identification process. Then, we test this architecture over a noise. Using the same technique, we compute the neural architecture to control a deadband system. The control loop is based on the internal model control theory.
[1] C. Desoer,et al. Stability of dithered non-linear systems with backlash or hysteresis , 1986 .
[2] Y. Autret,et al. Using SIGNAL for developing neural control systems , 1994, Proceedings of OCEANS'94.
[3] Carlos E. Garcia,et al. Internal model control. A unifying review and some new results , 1982 .