采用BP神经网络拟合光纤位移传感器特性曲线

Using BP neural network to realize optic fiber displacement sensor's curve-fitting

采用BP神经网络对光纤位移传感器特性曲线进行拟合。通过实验获得光纤位移传感器的实验数据,引入BP神经网络对光纤位移传感器的输入输出特性曲线和迟滞特性曲线进行拟合。Matlab仿真结果表明,该方法取得了较好的曲线拟合效果。

This paper adopts BP neural network to realize optic fiber displacement sensor＇s curve-fitting. Firstly gets optic fiber displacement sensor＇s experiment data by experimental measurement, then introduces the BP neural network to fit input-out- put curve and hysteresis curve. So as to the performance of simulation by Matlab, it obtains better curve-fitting effect of optic fiber displacement sensor.