容忍性能退化的光纤光栅传感网络自愈机制及可靠性研究

Fiber Bragg grating (FBG) sensor network is a new optical sensing technology that has been widely applied to structural health monitoring (SHM). But the performance degradation of the FBGs will occur inevitably in the long-term service process when suffering from environmental erosion. To address the issues of abnormal demodulation, network failure and low accuracy caused by performance degradation of FBG sensor, this project attempts to propose an effective technique for FBG sensor network to enhance the fault-tolerance of distortion spectrum and self-healing capacity of failure sensor. In the project, the properties of distortional spectrum caused by performance degradation of FBG will be analyzed, and the dynamic demodulation model will be established using spectral reconstruction principle to achieve adaptive fault-tolerance of abnormal spectrum. By combining with the relativity analysis between FBG nodes and data extraction from the associated neighborhood FBG nodes, the online redundant model of FBG will be built based on extreme learning machine to replace the failure nodes, which will enable system to self heal in real time. To improve the accuracy and stability of measurement, a fusion of the demodulation model and the redundant model will be proposed using particle filter algorithm to optimize the healing process. Then, the reliability criteria will be designed for FBG sensor network to quantitatively evaluate the self-healing effectiveness under different degradation state. The project aims to enhance the fault-tolerant capability and self-healing function of FBG sensor network, and it will provide technical support for highly reliable and intelligent FBG sensor networks. Therefore the study is of great academic significance and reference value.

光纤光栅（FBG）传感网络作为结构健康监测领域新一代传感技术，在长期服役中依然不可避免受到环境侵蚀等因素影响出现性能退化现象。针对因FBG节点退化所引起的解调异常、网络失效以及精度劣化等问题，本项目力图为解决FBG传感网络畸变光谱容错和失效节点自愈提供有效的解决方案：分析FBG性能退化光谱的畸变特性，运用光谱重构原理构建高精度的动态解调模型，实现对异常光谱的自适应容错；结合FBG节点相关性分析，提取邻域关联FBG的传感数据，建立基于极限学习机的在线冗余模型以替代失效节点，完成传感网络的实时自愈；采用粒子滤波算法融合解调模型与冗余模型，优化自愈过程，提高传感精度和测量稳定性；设计FBG传感网络可靠性评价指标，量化评估不同退化状态下的传感网络自愈效果。本项目旨在提升FBG传感网络整体容错水平和自愈能力，为实现高可靠的智能FBG传感网络提供技术支撑，具有重要的学术意义与参考价值。

光纤光栅传感（FBG）网络在长期服役中不可避免会受到环境侵蚀等因素影响出现性能退化现象。针对因FBG传感节点退化所引起的解调异常、网络失效以及精度劣化等问题，本项目针对FBG传感网络畸变光谱容错和失效节点自愈模型设计开展研究，取得了以下研究成果：对FBG性能退化光谱的畸变特性进行分析与建模，构建了FBG畸变光谱重构模型，并基于该模型结合进化算法设计了高精度的动态解调模型，实现对畸变光谱的自适应容错；利用FBG反射光谱的时序特性建立神经网络解调模型，实现了对相邻FBG的重叠光谱的快速高精度解调；结合FBG节点相关性分析，提取邻域关联FBG的传感数据，构建了在线冗余模型，完成FBG传感网络的实时自愈；结合畸变光谱解调模型与FBG节点冗余模型，优化了FBG传感网络自愈机制；设计了FBG传感网络可靠性评价指标，量化评估不同退化状态下的传感网络自愈效果。本项目成果提升了FBG传感网络整体容错水平和自愈能力，为实现高可靠的智能FBG传感网络提供技术支撑，对进一步扩大FBG传感网络应用领域具有重要的参考价值。