The research of LVDT nonlinearity data compensation based on RBF neural network
Zhongxun Wang, Zhonghua Duan · 2008
This paper presents a method to compensate nonlinearity of linear variable differential transformer(LVDT) based on radial-basis function(RBF) neural network. Because of the mechanism structure, LVDT often exhibit inherent nonlinear input-output characteristics. The best approximation capability of RBF neural network is beneficial to this. We construct an self-adaptive neural network compensate system use the nonlinear fitting of the RBF network. The network training is most conveniently implemented using a gradient-decent algorithm and Gaussian function by importing the experiment data and the desired response. The simulation results show that the nonlinear compensation of LVDT based on RBF network models is effective and this is significative for the displacement measure.