campusreview. com. au workforce
We’ ve also used them to predict the behaviour of man-made slopes, such as road cuttings and open-pit mines, as well as ground anchors. The latter are used to stabilise structures such as transmission towers, road cuttings, rock faces and so on.
These methods you’ ve mentioned have been recognised as allowing engineers to design cheaper and safer civil infrastructure across the globe. What does this mean for the industry? To start off with, we should mention that geotechnical engineering problems are particularly complex because we cannot choose the material we have to work with, and it is often difficult to determine all of its key mechanical properties accurately.
It’ s not like a concrete structure or a steel structure, where engineers can pre-determine the material. The methods I have developed require only strength parameters, which engineers can generally measure fairly reliably from laboratory and field tests.
As I mentioned, they give both upper bounds and lower bounds on the failure load of the infrastructure. That means engineers can estimate the true failure load with increased confidence.
Because the methods are extremely fast, typically requiring less than a minute or so to run on a PC, engineers can use them to investigate a wide range of design scenarios quickly, which means more cost-effective and safer designs.
Can you talk us through some of the other key breakthroughs you’ ve been involved with over the years? Yes. Mostly in the field of what’ s known as computational geomechanics. This has involved the development and application of computer methods to solve practical engineering problems. I was the first to show you can use high-order elements to predict failure with conventional finite element methods. These elements are now used in a variety of commercial codes, such as Plaxis, which is one of the world’ s most commonly used computer programs for geotechnical analysis, as well as in a variety of research programs at universities such as Newcastle, the University of Western Australia, the University of Sydney, Purdue University and MIT in the US, and Cambridge and Oxford in the UK.
I have also worked on new methods for implementing advanced soil models in finite element programs. Soil models are often complex, compared with most models we use to predict the behaviour of materials. These new methods are theoretically rigorous and control the error in the computed solutions for stress and deformation that engineers need to design infrastructure. The resulting algorithms have been implemented in a wide range of commercial computer programs, as well as in research programs at places such as Imperial College, Cambridge and Oxford in the UK, and MIT, Berkeley and Purdue in the US.
I’ ve developed new methods for computing finite element meshes. If you’ re going to do a finite element analysis, you need the mesh. I’ ve also developed new methods for solving large, sparse systems of linear equations. These arise everywhere in finite element methods, and again my methods and computer codes are used widely in industry and research across the globe.
You are the director of the Australian Research Council Centre of Excellence in Geo-technical Science and Engineering. What changes in the field have you seen during your time in this position and what are the biggest changes the industry is seeing at the moment? I think the biggest change we’ ve seen is that there has been rapid growth in the use of quite sophisticated computer methods in practical engineering design. This has posed a major challenge for engineering firms, as their staff need to be educated in not only the use of these methods, but also their limitations and underlying assumptions.
It should be said that the mathematical theory that underpins many of these techniques is often complex and subtle, and substantial amounts of training are required to get on top of it. I think the other big challenge in engineering, and in my field of work in civil engineering, is that many of the problems we deal with are now multidisciplinary, often involving areas such as chemistry, physics, biology and applied mathematics. Engineers need to have, at least, a fairly solid knowledge of some of these areas in order to make informed decisions about designs that are acceptable to the community.
What are the main influences behind this? I think the main influences are the relentless drive for more cost-effective solutions to engineering problems and the rise in community expectations. I think there’ s also a renewed focus on the importance of infrastructure to the national economy – not only in Australia, but also worldwide. Hence, we now have governments of various political persuasions claiming they are infrastructure governments and so on.
This trend is not going to go away, and infrastructure is going to be a problem that’ s going to be with us for quite a while. Just to cite a number, for example, it’ s been quoted that traffic congestion in Sydney alone costs about $ 5 billion annually to the New South Wales economy.
Finally, what do you hope to work on next and where do you see the industry moving over the next five years? The ARC Centre of Excellence for Geotechnical Science and Engineering is focused on energy and transport infrastructure, and I’ ll continue to work on a variety of problems in that field.
Physical infrastructure, such as national road and rail systems, and offshore facilities and pipelines, rely on geotechnical engineering design and this is increasingly associated with building on soft ground.
Onshore, we have transport corridors that must frequently make use of poor soils that have proved problematic for other developments, while offshore, soft sediments are encountered in almost all our recent developments where water depths now mostly exceed half a kilometre.
We’ re increasingly going out to greater depths in the ocean, and onshore we have a lot of problems with soft soils, and we need cost-effective solutions. The problem for engineers is that in all these cases the response of the material we’ re trying to build the infrastructure on, whether it is an offshore platform or a road or a rail system, is complex, highly variable and poses major design challenges.
Further to infrastructure on soft soils, in September 2016, the ARC Centre of Excellence will be hosting an embankment prediction symposium in Newcastle. This will allow engineers from all over the globe to test their predictive skills against data collected from a trial embankment that has recently been completed at Ballina in northern New South Wales. This embankment is fully instrumented, using state-of-the art equipment, and will provide engineers with a rare opportunity to observe the detailed behaviour of a full-scale geotechnical structure on soft soil.
The aim of this exercise is to improve upon the safety and cost-effectiveness of geotechnical design procedures for the construction of infrastructure, such as road and rail corridors, on soft soils. n
33