Nonlinear Compensation Method for Transducer Based on Neural Network with Fourier Functions
Nan Lan Wang, Ming Shan Cai · Applied Mechanics and Materials · 2012
Aiming to solve the problems in the non-linearity of thermistor temperature transducer, a compensate model based on neural network (NN) is proposed. The basic idea is using Fourier series as the basic functions of NN，the output of transducer as input samples of NN and the temperature as the expectation output of NN. The output of NN is used to approximate to the measured temperature by adjusting the weights. The results show the proposed method is effective in raising accuracy.