Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Civil-Comp Proceedings
ISSN 1759-3433
CCP: 78
Edited by: B.H.V. Topping
Paper 53

Development of AI Techniques for the Condition Monitoring of Ground Anchorages

A.J. Starkey, A. Ivanovic, R.D. Neilson and A.A. Rodger

Engineering Department, University of Aberdeen, United Kingdom

Full Bibliographic Reference for this paper
A.J. Starkey, A. Ivanovic, R.D. Neilson, A.A. Rodger, "Development of AI Techniques for the Condition Monitoring of Ground Anchorages", in B.H.V. Topping, (Editor), "Proceedings of the Seventh International Conference on the Application of Artificial Intelligence to Civil and Structural Engineering", Civil-Comp Press, Stirlingshire, UK, Paper 53, 2003. doi:10.4203/ccp.78.53
Keywords: artificial intelligence, neural networks, signal processing, geotechnical engineering, ground anchorages.

The GRANIT system is a non-destructive method for the condition monitoring of ground anchorages. Ground anchorage systems in the format of ground anchorages or stratabolts are used extensively throughout the world as supporting devices for civil engineering structures such as bridges and tunnels. A need for the condition monitoring of ground anchorages has been identified [1], with only between 1-5% of anchorages currently being monitored in service [2].

The GRANIT system operates by applying an axial load to the anchorage by way of a specially designed impulse device that connects to its protruding length. The acceleration signals are recorded by a laptop computer or equivalent, and the signals are analysed by signal processing techniques for the detection of relevant characteristics in the signal that relate to specific features of the anchorage (for example, its load or the size of its free length).

This paper describes recent work undertaken at AMEC's specially designed all weather stratabolt test site at Swynnerton, England, for the investigation of the applicability of the GRANIT technique to ground anchorages or stratabolts that are used throughout the UK Coal Mining industry. Results from these trials will be shown, showing the high accuracy in load detection that can be obtained. The Artificial Intelligence schema employed by the GRANIT technique will be described fully and the novel introduction and implementation of a mathematical model of the anchorage system into this schema will also be explained.

The development of the AI schema from its original to the present form will be described, showing how various AI schemes have improved and added to the knowledge base of the condition monitoring system. The full process of gathering data from anchorages using the GRANIT system to the training of neural network systems for the diagnosis of the load of these anchorages will be described, highlighting the difficulties encountered at this new all weather stratabolt test site, and how these problems were overcome. This has resulted in a new method for the diagnosis of load that utilises the mathematical model, and how this has resulted in the superior diagnosis of load (i.e. diagnosis with a smaller error than for neural networks trained on data taken from the anchorages).

The implementation of the mathematical model has implications for the diagnosis of other characteristics of an anchorage system, and the additional functionality of the GRANIT system in this regard is also discussed with the an example of how the free length of the anchorage system can also now be diagnosed following these investigative trials at the all weather stratabolt test site.

ICE,"A National Agenda; Long-term & Fundamental Research", Institution of Civil Engineers, Thomas Telford, London,1992.
BSI, "BS8081 British Standard Code of Practice for ground anchorages", 1989.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £82 +P&P)