提出了一种新颖的、采用周期性对称与不对称刻纹多模光纤组成的传感阵列网络,可用于智能材料与结构的应力、应变等的状态探测或损伤估计。探索采用了人工神经网络技术处理其阵列传感信号,给出了BP网与Kohonen网的仿真结果。

Presented in this work is a novel fiberoptic sensor array which is made of symmetric and asymmetric periodically etched multimode optical fibers. It may be applied to detection of structural states such as stress, strain or damage assessment in smart material and structures. The principle and technology of array sensing signal processing in the artificial neural network are described. The simulation results of BP network and Kohonen network are also given.