Information Technology Journal1812-56381812-5646Asian Network for Scientific Information10.3923/itj.2014.1161.1167ChuChao-Ting ChiangHuann-Keng 62014136This study proposed the implementation of adaptive sliding mode recurrent Gauss basis function neural network estimation in magnetic bearing system. The magnetic bearing system is very unstable nonlinear systems, so that the nonlinear controller is suitable to have a good response. In the traditional sliding mode control, the sign function produces chattering phenomenon and if the sign function is replaced by the saturation function which makes a steady-state error output. Hence, sliding mode control with neural estimation was proposed in this paper, in which the neural estimator improves the chattering phenomenon and steady-state error. This study uses the simple structure single-input single-output of the recurrent Gauss basis function neural network which reduces the system computation and has good estimation effect. The lumped bounded uncertainty E is a linear combination of the weights between hidden layer and output layer in which recurrent Gauss basis function neural network has better accurate estimation value than the Gauss basis function neural network. Hence, the output responses in the recurrent Gauss basis function with sliding mode control are better than the sliding mode control.]]>Chen, S.Y. and F.J. Lin,201119636643Bangcheng, H., Z. Shiqiang, W. Xi and Y. Qian,20124819591966Fang, J., J. Sun, H. Liu and J. Tang,20104640344045Bachovchin, K.D., J.F. Hoburg and R.F. Post,20124821122120Lin, F.J., K.K. Shyu and R.J. Wai,20016453466Lin, F.J., S.Y. Chen, K.K. Shyu and Y.H. Liu,20105716261640Lin, F.J., S.Y. Chen and K.K. Shyu,200920938951Li, X. and W. Yu,20102010pp: 615619Schweitzer, G. and E.H. Maslen,2009Pages: 535Pages: 535Narendra, K.S. and A.M. Annaswamy,1988