Back to search

FORNY20-FORNY2020

KVAL: Development and application of distributed acoustic sensing (DAS) in connection with critical infrastructure

Awarded: NOK 0.50 mill.

Project Manager:

Project Number:

342452

Project Period:

2023 - 2023

Funding received from:

SFI Center for Geophysical Forecasting (CGF) at NTNU in Trondheim, has carried out several tests using existing fiber optic communication cables to monitor critical infrastructure. The specific method is based on "Distributed Acoustic Sensing" (DAS) and is a technique where pulsed laser light is sent from an instrumentation connected to the fiber such as along critical infrastructure. An acoustic wave will cause a small stretch, or compression, in an optical fiber. In a telecommunications fiber there are small points of scatters, which are variations in the refractive index, which then reflect the laser light back through the fiber. The position of these spreaders is affected by compressions and stretches. The backscattered laser light is then converted to phase changes that are a function of the acoustic/elastic energy exerting a force on the cable/fiber. The changes in phase of the laser light are a function of the strength of the acoustic wave. The fiber is then synthesized into a series of receiver points that can be separated down to approx. 1 m and can measure any vibration close to the fiber with potentially high resolution in space and time. This has triggered a significant potential for utilizing existing fiber optic networks as distributed sensors to investigate events and changes in various environments, especially in connection with critical infrastucture. The sensor network is already laid out and existing, and it is possible to utilize the available network for various monitoring applications. The project is divided into two parts, one part dealing with market and business investigations and one part with testing of DAS measurements with an instrumentation connected to an available communication fiber in Bane NOR's telecommunications room at Støren. The measurements were conducted on the railway between Støren and Drivstua on Dovrebanen. The outcome of the project provide additional information as part of the process to create a DAS service company as spin-off from NTNU. The project will influence the strategy of how to define the area of applications and the customers prior to start-up and during the early phases of the coming spin-off company. Phase 1 of the project was a study to identify the main areas where distributed acoustic sensing could make a difference in monitoring of critical infrastructure. The offshore energy distribution segment, and the onshore power-line infrastructure (general power supply market) appear to be the most attractive areas where DAS-technology will find applications. The volumes and the potential capability of payment for the services among these market operators are larger than for instance railway, roads and security segments. In a competitive market, it is necessary for a start-up company to develop and offer AI-driven data processing to perform detection and classification of events and effective display pertinent alerts in connection with infrastructure monitoring. It is also imperative that the data analysis and alerts messages can be adjusted to sector specific requirements. Phase 2 of the project consist of two periods of DAS measurements utilizing an idle (dark fiber) along the railway between Støren and Drivstua at Dovrebanen. One of the main goals of the project was to test if there was a possibility to observe larger wild animals on, or close to, the railway prior to any collision. However, the number of collisions reported were too few to make any conclusion of this specific task. In addition, some of the reported collisions between trains and wild animals were outside the range of the specific measurement area. The specific task will have to be further investigated through longer term measurements to add up enough statistics data analysis. An important aspect of the test along the railway was to fine tune the acquisition parameters for applications to railway monitoring. Detecting moving trains is an obvious added application, including having the possibility to distinguish between different types of trains. This is possible by analyzing the strength and signature of the received acoustic signal. The DAS online monitoring also appear to be closer to real-time than the Bane NOR train tracking to be found at their home page. Construction work on the railway was also occasionally detected. The acoustic "tap-test" carried out at specific areas along the railway showed that various kind of signals, such as jumping and stomping by humans, as well as using a hammer to tap the side of the rail, were also picked up by the DAS instrumentation. Furthermore, it is evident that the DAS-measurements will be able to detect other possible events such as rock falls, snow avalanches, "wheel flats" and possible sabotage if the unwanted events and activity will involve emission of acoustic signals. DAS monitoring along railways will for sure contribute to better safety, increased operational efficiency and improved maintenance over time.

The Phase 1 of the project was a study to identify the main areas where distributed acoustic sensing could make a difference in monitoring of critical infrastructure. The offshore energy distribution segment, and the onshore power-line infrastructure (general power supply market) appear to be the most attractive areas where DAS-technology will find applications. The volumes and the potential capability of payment for the services among these market operators are larger than for instance railway, roads and security segments. In a competitive market, it is necessary for a start-up company to develop and offer AI-driven data processing to perform detection and classification of events and effective display pertinent alerts in connection with infrastructure monitoring. It is also imperative that the data analysis and alerts messages can be adjusted to sector specific requirements. The Phase 2 of the project ended up in two periods of DAS measurements utilizing an idle (dark fiber) along the railway between Støren and Drivstua at Dovrebanen. The main takeaway from this measurements were to fine tune the acquisition parameters for applications to railway monitoring. In addition, it also showed that the maximum possible distance to monitor by the use of the specific DAS-instrumentation was a distance of 100 km. The obvious events to measured during the test periods were moving trains and that it is possible to distinguish between various trains (person traffic and goods) by just analyzing the impact and signature of the online display. This online monitoring appear to be closer to real-time than Bane NOR home page train tracking. Construction work on the railway was also an obvious event to be detected. One of the main goals of the project was to test if there was a possibility to observe larger wild animals on, or close to, the railway prior to any collision. However, the number of collisions reported were to few to make any conclusion of this specific task. In addition, a some of the reported collisions between trains and wild animals were outside the range of the specific measurement area. So, this specific task will have to be further investigated through longer term measurements and covering a larger area than just the 100 km to add up enough statistics for data analysis. The acoustic "tap-test" performed at specific areas along the railway showed that various kind of signals, such as jumping and using a hammer to tap on the side of the rail track, were easily detected by the DAS-instrumentation. Furthermore, it is evident that the DAS-measurements will be able to detect other possible events such as rock falls, snow avalanches, "wheel flats" and possible sabotage if the unwanted events and activity will involve emission of acoustic signals. DAS monitoring along railways will for sure contribute to better safety, increased operational efficiency and improved maintenance over time.

Funding scheme:

FORNY20-FORNY2020