A rising demand for underground transportation and resource management has led to the development of many more subterranean projects (deep foundations, tunnels, utility corridors, etc.), which are constructed at larger scales, over greater distances, to increased depths, and within proximity to sensitive urban environments (i.e., reduced tolerances with respect to adjacent infrastructure). The result in any given project is a wide variety of ground mass and in-situ stress conditions that will often require some form of ground support or reinforcement.
For such projects, engineering design of such support is primarily based on the stress and strain that are developing within the support structures as a result of the surrounding ground conditions. These ground loads are distributed continuously and spatially and as such, an improved understanding of the continuous strain profile would provide better insight into the true behaviour of such support elements.
Research currently being conducted at the Royal Military College of Canada (Figure 1) focuses on such micro-scale geomechanical mechanisms and interactions with a view to determine the overall design implications for full-scale support design for tunnels (for example).
Use of Fiber Optics
The use of fiber optics within the geotechnical and geological engineering field is not a new concept. There are multiple projects that have utilized a particular type of fiber optic technology in the past, ranging from their use to monitor the construction and performance of embankments, tunnels, piles, mining operations and other geotechnical works (Shi. B., et al. (2006), Kalr. A., et al. (2006), Naruse, H., et al. (2007), Cheung. L.L.K., et al (2010), Lu, Y., et al. (2012)). It is important to note that not all fiber optic technologies are similar as each type has its unique strengths and limitations. Several methods of optical strain sensing are currently available that use standard singlemode- fiber (SMF); a low-cost telecom fiber. These include technologies that are Raman, Brillouin, and Rayleigh scattering based, as well as Fiber Bragg Gratings (FBGs) based techniques.
Of interest to the authors are the specific behaviour and interactions of ground support elements (i.e., rock bolts, spiles, cables, forepoles, etc.) used within underground civil construction and mining activities. As cited by Hyett et. al., (2013), measuring the strain (i.e., load) along fully grouted rock bolts remains appealing and much valuable research has been conducted using foil resistance strain gauges, particularly in US coal mines. However, more extensive implementation of this type of instrumentation remains unlikely because:
- The technology is expensive;
- The technology is perceived as challenging to successfully implement due to the lack of skilled experts;
- The reliability of the technology is questionable within harsh underground environments; and,
- The technology relies on discrete (pointwise) pairs of strain gauges that may not capture the complete strain profile along the bolt.
As such, the research and development that is being conducted at the Royal Military College of Canada alongside our corporate sponsor, YieldPoint Inc., offers a fiber optics sensing technology solution that can address each of the abovementioned limitations.
In the initial stages of the research, a laboratory testing scheme was developed and conducted with a view to capturing the combined performance of ground support elements. Central to this research is the application of a novel distributed optical strain (DOS) technology in combination with various ground support elements. Historically, monitoring of such ground support members has been limited to electrical and mechanical techniques (i.e., foil-resistive strain gauges, inclinometers, linear variable displacement transducers).
Such techniques provide discrete measurement points, implying that many sensors are required to obtain a full strain profile along the length of the support element. Such arrangements can be seen in Figure 2.
As can be seen, these techniques provide a limited spatial resolution along the element, making such methods prone to misinterpretation, underestimation, and possibly omission of support response. For example, it is not uncommon to observe a ‘failed’ rock bolt that has been subjected to both axial loads as well as bending (i.e., transverse loadings). As depicted in Figure 3, a coarse monitoring arrangement could potentially miss such local mechanisms completely.
Optical Fiber Technology, Specimen Preparation and Methodology Developed
An optical frequency domain reflectometry (OFDR) technology using low-cost, single mode optical fiber was investigated as a potential distributed strain monitoring technique for ground support members. The optical technology monitors spectral shifts in the local Rayleigh back-scattering arising from alterations in the index of refraction of the optical fiber, which is inherently sensitive to strain. In this manner, an initial measurement can be recorded in order to reference all succeeding measurements and therefore determine changes in strain.
What makes this OFDR technology particularly attractive for monitoring the aforementioned support elements is the capability to monitor strain with a spatial resolution of 0.65mm along the length of the optical fiber sensor. As well, the operational accuracy is quite acceptable (better than +/- 10 microstrain). The analyzer that was utilized for this research was Luna’s ODiSI-B unit.
In essence, thousands of individual transducers can be replaced with a single optical fiber acting as both the transducer and the lead. Thus, by equipping such an optical fiber along the length of a support member a continuous strain profile can captured.
No. 6 Grade 60 rebar were prepared by instrumenting them with Rayleigh DOS in order to conduct the laboratory experiments. This was done according to the methodology outlined by Forbes (2015).
All steel bars were modified with 2.5mm by 2.5mm diametrically opposing grooves as shown in Figure 4. This corresponds to an overall reduction of the cross sectional area of the rebar of four per cent.
These slots provided more than enough space for the application of the optical fiber. The modified rebar had a theoretical altered yield and tensile strengths of 117.3 kN and 183.1 kN respectively.
Prior to coupling the fiber optic sensor to the surface of the steel bar, the sensor had to be prepared. The sensors were made up of a single mode optical fiber, a terminating end, and a connector end.
The fiber was installed as a “bare” optical fiber (i.e., with a core, cladding, and buffer layer) in order to ensure optimum coupling between the steel surface and the fiber itself.
A lucent connector (LC) was used in order to connect the fiber optic sensor to the Luna ODiSI-B unit. Once the optical sensor was ready to be coupled to the steel bar, the machined grooves had to be properly prepared for instrumentation.
This proceeded by surface abrading the grooves with 220 grit sandpaper and cleaning them with acetone. The fiber was then placed within the prepared grooves and tensioned.
The sensor was looped at the toe end of the rebar through a machined out slot.
Once the optical sensor was positioned in the machined grooves, the precise location of the fiber along the alignment of the rebar was carefully logged according to the placement of the fiber along the grooves.
Selected Laboratory Testing and Results
To date, many configurations of testing that include axial, bending, and shear testing have been conducted utilizing multiple support elements.
These support elements were tested as unique specimens as well as grouted within concrete (rock) samples. The support tested in the laboratory to date includes: Rebar (rock bolts);
- D-bolts;
- Cable bolts;
- Spiles; and,
- Forepoles
Each sample preparation has its own unique challenges in terms of fitting the fiber optics in conjunction with a particular support element.
Also developed during this stage of testing was a fiber optic probe. The technique considers monitoring three sensing lengths along the profile of a fully-grouted rock bolt element using a single optical fiber, which, in turn, allows the derivation of both the principle strain and principle strain direction along the bolt. This is outlined by Forbes et al, 2017.
In Figure 5 one can see selected results from the laboratory tests that have been conducted as part of this line of research. Figure 5a depicts results from an axial pullout test while Figure 5b depicts results from a 2-way shear test.
Field Trials
As with any technology of this nature, it is encouraging to obtain excellent results within the controlled environment of the laboratory. The question now becomes how this technology can be employed in the harsh conditions associated with the field while limiting its impact on operations.
To date, multiple successful field experiments have been conducted at three separate locations around the world.
The authors are also in contact with other interested global parties who have shown an interest in employing such a technique within their operations.
Figure 6 (next page) is a series of relevant photos from the in-situ installation of the fiber optic technology within support elements that were designed by the authors. The data amassed in the field to date is of excellent quality, however, at the time of publication these results had not been authorized for release.
Nonetheless, it is extremely encouraging that the technology developed and tested at RMC is functioning as expected within the austere field site conditions with no real interruption to tunnelling or mining operations.
It should also be noted that a unique fiber optic instrumentation solution must be determined for each type of ground support element; this is a non-trivial undertaking due to the unique requirements and installation procedures associated with each type of support and site.
Conclusions
The distributed optical strain sensing technique has been verified as a novel monitoring and geotechnical tool for capturing the performance of ground support members used in underground projects.
The sensitive spatial resolution allows a continuous strain profile to be measured, overcoming the limitations of conventional, discrete strain measuring techniques, which in most cases will not fully capture the geomechanical behaviour of the support, especially when considering localized complexities.
The results of using this instrumentation with ground support elements in the laboratory and the field have provided confidence for using and improving upon such a technique within the field.
In addition, the optical technique can be realized as a novel tool with the capability to “see” and “sense” into the ground ahead of the working face, allowing the engineer to react and make adjustments to the support and excavation process in response to future ground conditions.
As a monitoring solution, DOS provides unparalleled information concerning the behaviour and the interaction between the ground medium and the support elements which can be back-analyzed for predictive numerical model methods and ultimately support design optimization.
