Examining the influence of TBM-ground interaction on electrical resistivity imaging ahead of the TBM
Introduction
Tunnel boring machine (TBM) excavations are increasingly performed in urban environments (underneath critical buildings, beneath roadways and utilities, in close proximity to other tunnels and foundations) and in more complex geologies (mixed face conditions, water table, karstic rock, etc.). Under such conditions, the excavation process incurs higher risk due to an unanticipated presence of ground changes, water bearing zones, weak regions or voids, and undocumented structural remnants (tie back anchors, abandoned wells, etc.). A considerable limitation in current TBM tunneling practice is the inability to characterize these unexpected changes ahead of the TBM. Methods that can continuously, and in real time, predict geologic changes and other hazardous conditions ahead of the TBM would provide significant risk reduction during excavation.
The field of geophysics offers non-destructive and continuous techniques that can potentially image ahead of the TBM without halting excavation like other more destructive methods such as probe hole drilling. At least two TBM-integrated-electrical resistivity systems are commercially available, the BEAM (Kaus and Boening, 2008) and the BEAM4 (Kopp, 2012), and have been implemented on a few projects worldwide. Still, publications that document the underlying principles, implementation details, as well as strengths and limitations are scarce. Electrical resistivity has successfully been used in a multitude of applications including vadoze zone hydrology, oil and gas location, pollution detection and ore body mining (e.g., Mazac et al., 1987, Benson et al., 1997, Meju, 2000, Atekwana et al., 2000, Atekwana et al., 2002). Given its success in other applications, is it unclear why electrical resistivity has not yet been fully developed and adopted universally in TBM tunneling.
The goal of this paper is to explore the fundamentals of TBM-integrated electrical resistivity and examine the complexities associated with TBM excavations and their influence on TBM-integrated electrical resistivity. Surface-based electrical resistivity is first presented to provide the fundamental basis for its application to the TBM tunneling environment. The paper then defines the important aspects of the TBM tunneling environment for both soft ground pressurized face and hard rock open mode conditions, particularly the TBM-ground interface and material properties within the interface. Six unique TBM-integrated-electrical resistivity electrode configurations are considered to determine how each configuration is influenced under different interface conditions. The paper will examine many material and geometric combinations for each of the TBM-integrated-electrical resistivity electrode configurations by using computational finite element modeling. This paper focuses on so-called DC resistivity methods and does not consider complex conductivity observed at higher AC frequencies through induced polarization (IP) methods. The extension to complex conductivity and IP is discussed in the paper.
Section snippets
Electrical resistivity background
An analysis of surface-based electrical resistivity provides the needed background for the TBM tunneling framework. A traditional surface electrode array is comprised of four electrodes A, B, M and N (Fig. 1). DC or very low frequency (<1 Hz) AC current is injected between electrodes A and B, and the resulting electric field (measured as an electrical potential in Volts) is measured between electrodes M and N. Although higher frequency AC current injection has been adopted more recently to
TBM tunneling environment and the interface region
The TBM tunneling environment is modeled with a hollow cylinder of diameter D located at a cover depth of C below the earth’s surface (Fig. 3(a)). The TBM structure is located at the closed end (right-hand side in Fig. 3) of the cylinder and is comprised of both the cutterhead and the shield. To the left of the TBM structure is a lining constructed of either shotcrete, cast-in-place concrete or precast concrete segments. Unlined tunnels are not considered here, but are of no consequence to the
Finite element model
Finite element models were developed to perform the analysis in the following sections. The models use the COMSOL Multiphysics finite element software. COMSOL solves physics based partial differential equations to observe physical phenomenon in three dimensional space. COMSOL’s AC/DC module simulates the flow of electrical flow through three dimensional space (e.g. surfaces, volumes, points) given user defined geometry, electrical material properties (namely electrical conductivity), and
Current density analysis
Current density contour plots convey how well each electrode array can inject current ahead of the TBM for each of the various interface region geometries and material types. Fig. 8 shows contour plots of current density (mA/m2) on the vertical x-z plane (y = 0) around the TBM cutterhead for the electrode configuration shown in the center of the plots (configuration from Fig. 5(a)) and for hard rock tunneling where the interface region is air (σA = σD = 10−15 S/m). Each of the four contour plots in
Limitations and implementation issues
This study investigated only two combinations of electrical conductivity for the formation around the TBM (σ1) and the vertical planar difference (σ2). In reality, a wide range of electrical conductivities could be present in much more heterogeneous formation. While the general trends presented above remain true, the levels of current density and potential measurement will change with electrical conductivities. Noise is an important consideration for electrical resistivity methods. It is not
Conclusions
This paper presents finite element results from a three dimensional study on electrical resistivity methods applied to the TBM tunneling environment. We focus on identifying the electrical influence of a volume that interfaces the TBM to the virgin formation, called the interface region. There is a degree of uncertainty regarding the interface region regarding its electrical conductivity and geometry and so this study considers a range in each. Six electrode arrays are considered that are
Acknowledgements
We thank NSF for funding the SmartGeo Educational Program (Project IGERT: Intelligent Geosystems; DGE-0801692). The opinions expressed in this paper are those of the authors and not the NSF.
References (25)
- et al.
Investigations of geoelectrical signatures at a hydrocarbon contaminated site
J. Appl. Geophys.
(2000) Performance prediction of TBM disc cutting on granitic rock by the linear cutting test
Tunn. Undergr. Space Technol.
(2006)- et al.
Surface geoelectrics for groundwater pollution and protection studies
J. Hydrol.
(1987) Geoelectrical investigation of old/abandoned, covered landfill sites in urban areas: model development with a genetic diagnosis approach
J. Appl. Geophys.
(2000)Electrical resistivity survey in soil science: a review
Soil Tillage Res.
(2005)Geophysical investigation of vadose zone conductivity anomalies at a hydrocarbon contaminated site: implications for the assessment of intrinsic bioremediation
J. Environ. Eng. Geophys.
(2002)- et al.
Mapping groundwater contamination using dc resistivity and VLF geophysical methods – a case study
Geophysics
(1997) - et al.
DC resistivity and induced polarization methods
(2005) Relationship between spectral induced polarization and hydraulic properties of saturated and unsaturated sandstone
Water Resour. Res.
(2005)- et al.
The influence of pore fluid chemistry on the induced polarization response of rocks and soils