When considering the cumulative effect of repeated seismic loading from earthquakes on tunnels one may not immediately think to speak with researchers at Leeds University’s Institute for Resilient Infrastructure. However this is exactly where world leading research into understanding tunnel behaviour is taking place. Under the “Shaking Tunnel Vision” project, fi nanced by the UK’s Department for International Development (DfID) Newton Fund, Associate Raul Fuentes is leading a two year project monitoring and modelling two major tunnels in Chile. “Data from the US geological survey website told us that there is a surprisingly huge number of earthquakes affecting Chile,” says Fuentes. “It revealed over 200 within 100km of Valparaiso with a magnitude higher than 4.0 and 17 higher than 5 only in a period of two years. It was an astonishing amount,” he says.
A road tunnel in the coastal city of Valparaiso will host one of the studies, with the other situated in a metro tunnel in the capital Santiago. At each location 50 instruments – 25 tri-axial accelerometers and 25 inertial measurement units will record the acceleration profi le and deformation over the two years of the study and will remain in the long term. The data will then be fed as a live stream to an online portal which will be analysed by the research team and made publically available.
The project also involves creating numerical models that will be validated as the data is collected so as well as improving understanding of tunnel behaviour under repeated seismic loading, the study will also enable the development of engineering risk based disaster management approach for tunnels in seismic areas. A third strand to the project is the creation of a new case study that will act as in internationally recognised reference point for improved design approaches.
“Design codes require that tunnels are designed to withstand large earthquakes but nobody has been interested in what happens to tunnels when you have multiple smaller earthquakes happening repeatedly. There is anecdotal evidence of cumulative damage of multiple earthquakes but it normally gets sorted out with smaller repairs and no one has done a systematic study,” says Fuentes.
Although Fuentes is the Principal Investigator, the University of Leeds is also working with the University of Dundee, the National University of Athens, ETH Zurich and the Pontificia Universidad Catolica de Valparaiso in Chile on the study.
Meeting on site in Chile in April enabled all parties, including the companies that manage the tunnels, to understand the research team objectives and there is huge interest in the potential findings which could, of course, could go either way. Tunnels may be so overdesigned already that the cumulative seismic effect is negligible over the lifetime of the structure. “But the answer may be that we observe some damage so then we will have to go back to square one in terms of what the codes say as at the moment they actually don’t say anything on cumulative effect,” says Fuentes.
At the same time the monitoring systems are set to give new data about critical assets and there is potential for the scope of work to expand further with over 200 tunnels in Chile that could benefit from monitoring systems.
LONDON CALLING
Imperial College London has been actively carrying out research into ground and structural response to tunnelling for more than twenty years through field monitoring, numerical analysis and model testing. Following major field monitoring campaigns on the JLE (Jubilee Line Extension) and CTRL the latest focus has been to investigate the response of existing tunnels to new tunnel construction. Field monitoring was used to assess the effect of building the new 7.1m diameter tunnels of the east-west Crossrail link on the existing Central Line underground railway tunnels. This major study was financed by the UK Engineering and Physical Science Research Council (EPSRC) with contributions from Crossrail Ltd, Morgan Sindall and collaboration with London Underground Limited. “It had five primary strings to it,” explains Principal Investigator Jamie Standing, Reader in Ground Engineering the Civil and Environmental Engineering Department of Imperial College London. “Field monitoring; structural testing on halfscale cast iron segmental rings; numerical analysis of field conditions; numerical analysis of the laboratory set-up and a comprehensive suite of tests on London Clay from high quality samples taken when installing the field instrumentation in Hyde Park. These strands linked together very closely,” he says noting that the numerical analysis was led by David Potts using the bespoke software ICFEP.
The existing Central Line tunnels were constructed from grey cast iron segmental tunnel linings, which are prevalent in many developed cities throughout the world. The results therefore offer lessons for many future schemes. “It taught us a lot,” says Mike Black, head of geotechnical engineering and lead on research papers for Crossrail Ltd. “We were assessing the ground interaction effect with this cast iron structure and understanding how that may have performed itself over time prior to Crossrail work, and subsequently the effect that Crossrail would have on that structure,” he says explaining that the combination of field measurements thorough extensive instrumentation, laboratory analysis and numerical analysis enabled these different aspects to be compared with one another.
“The findings should be very useful to future projects,” says Standing, noting that several papers have been produced from the study and others are still being written up. “Examples of key findings are that it is often thought that rings articulate but in the case of the Central Line tunnels just east of Lancaster Gate, where Crossrail ran 4.2m below the existing line, we found that all the longitudinal straining took place in the iron itself and not in the joints which is an important finding.” Another important aspect of the project was the study of bending moment distributions around the existing tunnel. “We have investigated carefully the Morgan Equation and Muir Wood’s modified version and have gained insight into how well they have worked from the structural testing that we did,” says Standing.
The approach used in this project is typical of how the Geotechnics Section at Imperial College works: validating numerical analyses with results from fieldwork and feeding in parameters determined from advanced laboratory studies. More recently there has been a focus on using the ground as a medium for storing thermal energy as a means of supplying sustainable, renewable and low-carbon heating and cooling to buildings. “Particular emphasis has been placed on the use of geotechnical structures, such as tunnel linings, as heat exchangers,” says David Taborda. “The associated temperature changes have the potential to generate additional ground movements, as well as structural forces, and are being investigated using newly commissioned laboratory equipment and newly implemented thermo-hydro-mechanical algorithms within ICFEP.”
Structural testing also features in the work of Nick Buenfield, head of civil engineering at Imperial College, who specializes in the long-term durability of concrete structures. Concrete-lined metro tunnels are particular interest as these are usually exposed to groundwater on the external face and drying conditions inside, causing aggressive contaminants in the groundwater to penetrate the concrete.
On the materials side Chris Cheeseman has been focussing on the reuse of waste materials, the circular economy and low-carbon cement and concrete, most notably discovering that heating waste London clay to temperatures of over 800 °C transforms it into a highly reactive pozzolan. This then has potential to be processed to form a technically and commercially viable supplementary cementitious material that could compete against ground granulated blast furnace slag and coal fly ash for use in concrete. Safety too has been a key research area with Tim Newman’s PhD research looking into causes of confined space hypoxia during underground construction in the Lambeth Group beneath London in conjunction with ground investigations for the Thames Tideway project. A major finding was that glauconite cannot cause the oxygen loss. Another mineral, green rust, has been identified as a potent and plausible reducing agent.
MEASURING MORE
Imperial College is not the only University to undertake research on London’s Crossrail project. At Cambridge research on the project began in 2009. “We had a research team working on structural monitoring of shafts and how they performed during construction, during excavation and the effect that it had on the ground around us. We did this through a knowledge transfer partnership set up by the Technology Strategy Board,” explains Mike Black.
The KTP was a vehicle that enabled Crossrail and Cambridge to work together. “That was a really excellent piece of work,” says Black explaining that the fibre optic sensors were installed in diaphragm wall panels. “It demonstrated that our structures were very robust, which in itself was very useful information. An incidental observation that came out of that work was the strain effects from changes in temperature.”
More recent research at the University has continued along the theme of monitoring full scale tunnels and comparing the monitoring data to the results generated by centrifuge modelling and numerical analyses. “If you understand infrastructure using sensors then you are in a better position to make data-driven informed decisions whether in the long term or in the short term when you have complex construction scenarios, particularly in projects where tunnels are constructed in close proximity to existing tunnels. These sensors will give you a full picture of how the tunnel(s) is behaving and allow you to be proactive as an engineer.” explains Mohammed Elshafie, lecturer at the Laing O’Rourke Centre for Construction Engineering and Technology at Cambridge University.
Elshafie points to two recent research studies on Crossrail that demonstrated the effectiveness of this approach. “In Bond St a new tunnel was being constructed right underneath the existing Royal Mail tunnel [at almost 90 degrees to it]. It was so close such that the new tunnel was touching the existing cast iron tunnel at the bottom of it. All risks were evaluated by the design and construction teams and mitigation put into place using conservative assumptions and rightly so.” A team from Cambridge University’s Centre for Smart Infrastructure and Construction (CSIC) went into the existing tunnel Royal Mail tunnel with a range of sensors from photogrammetry to fibre optic strain sensing gauges. “The level of detail and information that was extracted from these sensors during the complex construction process was unprecedented and hence, we now know a great deal about how these cast iron tunnels behave when you construct right underneath them,” says Elshafie.
Using the geotechnical centrifuge the team were then able to create a small scale model back at Cambridge and simulate the effects of the construction process using cutting edge centrifuge modelling simulation techniques. The results are currently being written up by three PhD students who will submit their theses in 2017. “The main trend that we are seeing is that the existing tunnels are more resilient than we think that they are. By understanding how they behave in a real scenario and combining this with detailed studies in the centrifuge, there is the potential that the industry in general can make some significant efficiency savings either by adopting appropriate construction methods and/or providing appropriate mitigation measures when needed. In general that gives UK construction industry a competitive advantage.”
Similar work was also undertaken beneath Liverpool Street but in this case the new tunnels were to run parallel to the Royal Mail tunnel. “It gave us a fantastic opportunity to look at two different scenarios. The behaviour is completely different; when it is perpendicular it is a localised effect, but when it is a parallel direction you get the effects along the entire length because you are digging beneath every single section. It was fantastic to see the difference between the two scenarios from the field data and even more fascinating to look at the models created (approximately. 1mx1mx1m) put into the centrifuge and replicate the same type of behaviour.”
Future areas of research at Cambridge are considering a range of topics including the redistribution of loads around cross passages and the logistics of tunnel boring using digital modelling which will enable clients, contractors to plan the tunnelling process in a virtual environment.
Comparing lab tests with real data is also a feature of the research at Nottingham University where a key research focus is tunnel and building interaction. “What we are doing in Nottingham is we are coupling numerical modelling and centrifuge modelling in such a way that we take advantage of the strengths of each modelling technique,” says Alec Marshall, associate professor at the Nottingham Centre for Geomechanics. “In a test, we replicate the effect of tunnelling on the soil and foundation system in the centrifuge and we pass information on foundation loads and displacements to a numerical model which solves for the redistribution of foundation loading based on the characteristics of the simulated building. The numerical model then passes back information about foundation loading to the centrifuge model so that the centrifuge model is continuously updated. This iterative process occurs very quickly, in near real-time, and continues until a specified level of tunnel volume loss is reached.”
He says that the results are giving us a better understanding of the realistic global interactions that occur between the building and tunnel construction. “As a result of this we are developing methods which will give a design engineer more efficient and effective tools for evaluating the effect of tunnelling on piled structures.”
IN THE CITY
London’s City University too is undertaking research work looking at the interaction between tunnels and existing structures. A current research project looks at the effect of tunnel excavation on escalator tunnels. Another focuses on ground support at the tunnel face and the effects on stability and surface settlement. Recently published work has looked at the impact of twin tunnelling. “Up until recently it was thought that the settlement above twin tunnels would simply be superimposed on one another. We did quite a big study that showed that actually you get more movement from the second tunnel. The effect of the first tunnel has a detrimental effect on settlement above the second tunnel,” explains Andrew McNamara, a senior lecturer in the civil engineering department of City University. At the same time the University has also been studying the disaggregation of slurries in pipe jacking, a project sponsored by the Pipe Jacking Association.
The PJA is also supporting a project at Portsmouth University which is studying the potential use of steel fibre reinforced concrete containing new hooked end steel fibres in jacking pipes. The objective of the project is to model the behaviour of SFRC pipes under proof loads using finite element modelling in order to develop suitable designs for various diameter pipes and to manufacture pipes according to these designs and test their performance experimentally.
SAFETY FIRST
Meanwhile at The University of Edinburgh fire safety is a key focus. For the past five years concrete spalling of precast lining segments during fire has been a major research topic, working in collaboration with consultant Arup. “When under high temperature gradients the concrete sometimes explodes away from the surface in layers, and that can be problematic for a whole host of reasons. So we are doing a lot of work at Edinburgh to try and understand that behaviour,” explains Luke Bisby, Arup Chair of Fire and Structures and Head of the Institute of Infrastructure and Environment, University of Edinburgh. “What we are trying to do is understand both the physical mechanisms that influence explosive spalling and then move on to mitigation strategies or design strategies.”
Fire safety expert Ricky Carvel also conducts research on tunnel fire safety at the University with his emphasis on ventilation rather than structural integrity. “I’ve spent two decades now looking at factors that influence fire growth in a tunnel. At the moment we are looking at the influence of ventilation on fire behaviour,” he says explaining that this really consists of a complex series of interrelationships between ventilation, smoke production, smoke movement and toxicity.
A recent research paper considered the implications for a fire in the centre of a passenger train in a twin tube tunnel, similar to the Channel Tunnel. By examining the implications of ventilating the tunnel in the event of such a fire the paper made the recommendation that not using mechanical ventilation was the safer option. Carvel now wants to further the research by looking at a more generalised scenario.
