• Mikhailov, Sergei

Abstract

Environmental and structural health monitoring is crucial for the effective and safe management of resources and assets. Said monitoring traditionally relies on electromechanical sensors, which may not always perform adequately when exposed to harsh conditions or they may simply not allow carrying out the required measurements in specific application scenarios. Sensor technology based on optical fibers can address these shortcomings and may hence be preferred over traditional electromechanical sensors in particular applications.<br/><br/>One of such applications is pressure measurements and sensing, which is of a great importance in oil and gas industry and in the field of geotechnics. This use case requires the ability to measure pressure in multiple points with a high spatial resolution over long distances. In the context of pressure sensing, several optical fiber point sensors have been developed. However, only a very few demonstrations of fully distributed fiber-based pressure sensors have been reported. A most likely reason for that is the negligible to small pressure sensitivity of conventional optical fibers, combined with their high cross-sensitivity to mechanical load and temperature.<br/><br/>In this thesis, we address this challenge, and we propose a solution by developing a distributed optical fiber sensor (DOFS) with enhanced sensitivity to hydrostatic pressure and that is only sensitive to pressure, i.e. the measurements do not need to be corrected for the effect of temperature changes. To do so, we depart from the use of conventional optical fibers and we develop a dedicated so-called ‘microstructured optical fiber’ (MOF) instead. Such a MOF consists of an all-silica optical fiber waveguide in which air-holes are introduced in the cladding region and extend in the axial direction along the entire length of the fiber.<br/><br/>The specific objectives of this thesis are: 1) to study to what extent pressure-sensitive MOFs are compliant with the existing DOFS techniques, 2) to investigate how distributed interrogating techniques can benefit from an optimized MOF design, 3) to design, model and fabricate such a novel MOF, and finally 4) to characterize and demonstrate the pressure sensing capabilities of our novel MOF.<br/><br/>First, we evaluate the feasibility of using Rayleigh scatteringbased DOFS methods for the distributed pressure sensing and we demonstrate – for the first time to our knowledge – the detection and location of fatigue cracks in welds in steel tubular specimens in large-scale fatigue tests using distributed strain sensing based on optical frequency-domain reflectometry (OFDR). We then demonstrate distributed pressure measurements based on a previously developed highly birefringent ‘Butterfly’ MOF, using frequency-scanned optical time-domain reflectometry. We show that this MOF features a pressure sensitivity up to ~8.8 times higher- than commercially available MOFs previously used for distributed<br/>pressure sensing. We then address the limitations of said Butterfly MOF in terms of distributed sensing, and we design and fabricate a set of novel birefringent MOFs with enhanced sensitivity to pressure and negligible sensitivity to temperature, which are adapted for distributed pressure sensing. Our MOF designs allows achieving a polarimetric pressure sensitivity sufficiently high to achieve the subbar pressure resolution and feature a record low cross-sensitivity to<br/>pressure and temperature. Finally, we characterize the distribution of the phase birefringence, as well as of the pressure and temperature sensitivities of our new MOFs using FS-OTDR. We demonstrate distributed pressure measurements with an unprecedented sub-bar pressure resolution and 1 m spatial resolution over a ~100m distance.<br/><br/>To conclude, we give an overview of future research perspectives for our sensor concept. We propose to look into the cabling of MOFs for use in harsh environments and we investigate how this would affect the waveguide and sensing properties of the fibers. We then point out the potential of using interrogation methods other than FSOTDR for distributed pressure sensing. We end with exploring the opportunities for distributed transversal strain sensing in smart<br/>structures using our MOFs.<br/><br/>We hope that this work impacts the field of distributed fiber sensing and contributes to the development and adoption of microstructured optical fiber technology as well.

Topics

• impedance spectroscopy
• phase
• crack
• steel
• fatigue
• reflectometry