Civil and Structural Engineering Computing: 2001
Chapter 18 H.R. Thomas, P.J. Cleall and D.H. Owen
Geoenvironmental Research Centre, Cardiff School of Engineering, Cardiff University, Wales Keywords: geotechnical, geoenvironmental, modelling, finite element method, stress-strain field
Increasing concerns relating to land use sustainability, waste disposal methods and environmental protection is currently making geoenvironmental engineering an area of significant international importance. A thorough theoretical knowledge is required for the solution of these problems combined with and insight into the fundamental principles of a) contaminant behaviour, b) control of pollutant fate, c) liquid and soil phase interaction in multi-phase systems and d) behaviour of both the waste and the interacting construction materials [7,8]. In particular, laboratory and field measured experimental results, can been used to validate and develop both conceptual and numerical models for use in geoenvironmental management. One geoenvironmental problem of importance internationally is the engineering performance of high-level nuclear waste disposal repositories. Deep geological repositories, where waste is placed within a multibarrier system employed to control radionuclides migration, are currently under consideration by several organisations (Swedish Nuclear Fuel Supply Co [2], DETR [1], EU 1999). The concept proposed involves the placement of a partly-saturated engineered clay barrier to form one layer of the multibarrier system. The behaviour of the system in relation to a) the flow of moisture, heat, gases and chemicals and b) the stress-strain response of the medium is therefore of great importance. Analysis of the bi-directional non-linear coupling between each of the flow fields and the stress-strain field is required. For example, the effects the system's chemical composition on the hydraulic and mechanical response need to be addressed. The authors and co-workers have been actively involved in the development of such models for several years. The prediction of the coupled movement of heat and moisture was addressed initially [5]. Varying air pressure was then included [6]. A coupled flow and deformation model, utilising a non-linear elastic constitutive relationship was then developed [3]. Later, the model was developed to include an elasto-plastic constitutive relationship [4]. This allowed collapse and swelling of the soil under wetting to be simulated. The next step was to attempt to include the effect of chemicals on the soil behaviour, extending the THM models to THCM. It should be noted that the work presented in this paper is a first attempt at representing this very complex problem. The theoretical approach developed and the numerical methods employed are described. Because of the computationally demanding nature of the work, attention has been paid to the implementation of parallel computing approaches and iterative solvers. The work performed in this context is described in some detail and the results achieved, in general terms presented. A validation exercise based on a coupled thermo-hydraulic-mechanical laboratory experiment is described. Work on the validation of the hydraulic-chemical component of the model is also presented with the model applied to simulate a series of experimental tests. The simulations presented show some encouraging results. Development of such geoenvironmental models is an ongoing process. In particular, further high quality experimental results are required to enable these developments to be rigorously validated.
References
| ||||||||||
