节点文献

大尺度光纤形态传感器关键技术研究

Research on Key Technologies of Large-Scale Optical Fiber Shape Sensor

【作者】 杨洋;

【导师】 沈平;

【作者基本信息】 哈尔滨工业大学 , 电子信息(专业学位), 2025, 博士

【摘要】 复杂环境中的大尺度形态测量技术在水下安防、灾害预警、能源运输等领域具有重要的应用价值。光纤形态传感技术凭借其非接触式测量、抗电磁干扰、耐腐蚀性强等特性,已成为柔性体三维形态重构的关键技术方案。然而,现有技术多聚焦于小尺度、大曲率测量(如医疗介入器械开发),对水下光电缆监测、建筑健康评估等工程场景中的大尺度(米级至公里级)形态传感仍存在理论方法不完善、技术手段匮乏等瓶颈问题。本研究面向大尺度形态测量需求,系统开展传感长度在1 m以上的大尺度光纤形态传感器关键技术攻关。针对弹性体约束下的大尺度形态传感模型构建、准分布式光纤光栅形态传感器研发以及用于形态传感的大动态高精度相干-光时域反射仪(Coherent-Optical Time domain Reflectometry,C-OTDR)应变传感技术研究三方面开展了理论研究和实验验证。本文的主要研究内容如下:第一,构建弹性体约束条件下大尺度形态传感理论模型。首次建立曲率传感单元布设密度与形态恢复精度的关系模型,通过Matlab数值仿真平台建模,系统性研究布设密度与曲率梯度分布的关联机制。构建分布式应变测量与形态重构的理论映射关系。通过力学本构方程分析,揭示约束体弹性模量与曲率测量误差的传递规律,推导出形态重构误差补偿公式,为米级至千米级形态测量提供了理论依据。第二,研制适用于深海环境的准分布式光纤光栅形态传感器并以此为核心传感单元组建海底滑坡监测系统。创新设计弹性增敏基体结构,提出差分曲率解耦算法,攻克多模态应变耦合干扰难题。光纤光栅形态传感器在平面单向弯曲测量中,豪斯多夫距离为6×10-3m,约为长度的0.20%,弯曲方向最大误差0.06 rad;在S形弯曲测量中,豪斯多夫距离为0.02 m,约为长度的0.77%,弯曲方向误差为0.04 rad。实验结果表明:光纤光栅形态传感器可进行高精度形态恢复,符合海底滑坡监测系统设计需求。基于形态传感器构建的海底滑坡监测系统在港池实验中实现全流程作业,为海底滑坡原位监测提供了新的解决思路。第三,研究了用于形态传感的大动态高精度C-OTDR分布式传感技术。针对C-OTDR应变测量中的幅度-频率谱畸变导致的相关性快速下降问题,提出了基于局部幅度-频率谱匹配的频移解调算法,降低了偏振衰落和非均匀应变导致的谱畸变对频移测量的影响。建立参考谱迭代跟踪的优化模型,将应变测量动态范围扩展至扫频限制的8倍。为C-OTDR分布式应变传感技术在大尺度形态测量领域的应用进行探索。

【Abstract】 Large-scale shape measurement technology in complex environments holds significant application value for underwater security,disaster early-warning systems,and medical devices.Fiber optic shape sensing technology has emerged as a critical solution for three-dimensional morphology reconstruction of flexible structures,benefiting from its non-contact measurement capability,electromagnetic interference immunity,and corrosion resistance.However,existing technologies primarily focus on small-scale/high-curvature measurements(e.g.,medical interventional devices),leaving theoretical and technical challenges in meter-to-kilometer scale engineering applications such as submarine cable monitoring and structural health assessment.This study centered on three pivotal research dimensions:1)spatial curvature sensor deployment modeling,2)quasi-distributed fiber Bragg grating(FBG)shape sensor development,and 3)high-dynamic-range C-OTDR strain sensing technology,the investigation combines theoretical modeling with experimental validation.The principal research components are structured as follows:Firstly,a theoretical model for elastic body-constrained shape sensing is established,proposing a shape reconstruction theory rooted in the Frenet-Serret framework.This work pioneers the development of a quantitative correlation model between curvature sensor deployment density and shape recovery accuracy.Through constitutive equation analysis,we reveal the error propagation mechanism linking the constraint body’s elastic modulus and curvature measurement deviations.A compensated shape reconstruction formula is derived.This model establishes theoretical foundations for cross-scale measurements spanning meters to kilometers.Secondly,a quasi-distributed fiber Bragg grating(FBG)shape sensor with deep-sea environmental compatibility has been developed,serving as the core sensing unit for submarine landslide monitoring systems.The innovation features an elastic sensitization matrix structure and differential curvature decoupling algorithm,effectively overcoming multi-modal strain coupling interference challenges.Experimental results demonstrate exceptional performance:In planar unidirectional bending,the Hausdorff distance is 6×10-3m(0.20%length)with maximum directional error 0.06 rad.In S-shaped bending,the Hausdorff distance is 0.02 m(0.77%length)with directional error 0.04 rad.The submarine landslide monitoring system has successfully demonstrated full-process operational capability in harbor basin experiments,establishing a novel methodology for in-situ detection of submarine landslides.Thirdly,we proposed a large dynamic strain measurement system of C-OTDR to address kilometer-scale shape reconstruction requirements.A frequency-shift demodulation algorithm based on local amplitude-frequency spectrum matching was developed,successfully overcoming critical technical bottlenecks caused by polarization fading and non-uniform strain-induced spectral distortions.An optimized reference spectrum iteration-tracking model was established,achieving a tenfold expansion in strain measurement dynamic range beyond conventional frequency-sweep limitations.This systematic investigation lays essential groundwork for applying C-OTDR distributed strain sensing technology to large-scale shape sensing.

  • 【分类号】TP212
节点文献中: 

本文链接的文献网络图示:

本文的引文网络