Wu, Jingyu
(2021)
Distributed Fiber Optic Sensors for Energy Applications.
Doctoral Dissertation, University of Pittsburgh.
(Unpublished)
Abstract
Fiber optical sensors are well known for their resilience in harsh environments. Since 2000, distributed fiber sensors have been explored for nuclear, fossil, and renewable energy applications. To improve the signal-to-noise ratio (SNR) and high-temperature stabilities, both the ultraviolet (UV) and femtosecond (fs) lasers have been used to inscribe in-fiber devices (fiber Bragg gratings (FBGs)) or enhance Rayleigh backscattering of pristine fibers to form so-called "sensing fibers."
This dissertation presents integrated research efforts in developing low-cost sensing fibers that have the potential to drastically improve applicability for fiber sensors as essential sensing tools for infrastructure monitoring, smart cities, and new energy applications. Both deep UV excimer lasers and fs ultrafast lasers have been used to form gratings and enhance Rayleigh backscattering of standard telecom fiber through protective coating to preserve the mechanical integrity and performance consistency of the fiber and improve the fabrication throughput for harsh environment applications. This dissertation research presents a detailed fabrication process and experimental results of distributed fiber sensors in a harsh environment: multiplexed FBG sensors and distributed fiber sensors with ~1.5-m fs-laser enhanced Rayleigh scattering centers withstood a total fast neutron fluence of 6×1020 n/cm2 in the core region. In-pile lead-out sensors tests were carried out at the MIT Research Reactor for two months, which was operated at a nominal power of 5.7 MW with a fast neutron (>0.1 MeV) flux of 1.29 × 1014 n/cm2/s and an in-core temperature of up to 560 °C. However, ionizing radiation-induced point defects and density fluctuations degrade fiber sensor performance by reducing SNR of fiber transmission and adding
v
measurement errors due to radiation-induced Bragg wavelength shift. To mitigate the complicated discrepancy in sensor performance due to the high radiation, a long short-term memory (LSTM)-based prediction model was applied to reconstruct and predict the distributed in-core temperatures at centimeter spatial resolution. The fiber optic sensor-based sensing system incorporated with advanced artificial intelligence (AI) technologies has shown the capability in online monitoring for next-generation nuclear power plants (NPP).
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
5 October 2021 |
| Defense Date: |
13 July 2022 |
| Approval Date: |
6 September 2022 |
| Submission Date: |
17 July 2022 |
| Access Restriction: |
2 year -- Restrict access to University of Pittsburgh for a period of 2 years. |
| Number of Pages: |
122 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
Swanson School of Engineering > Electrical and Computer Engineering |
| Degree: |
PhD - Doctor of Philosophy |
| Thesis Type: |
Doctoral Dissertation |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
distributed fiber sensors, harsh environment, artificial intelligence |
| Date Deposited: |
06 Sep 2023 05:00 |
| Last Modified: |
06 Sep 2024 05:15 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/43266 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}