MODELING HYGRO-THERMO-MECHANICAL BEHAVIOR OF ENGINEERED CLAY BARRIERS- VALIDATION PHASE

The Canadian Nuclear Fuel Waste Management Program (CNFWMP) is evaluat ing the concept of disposal of nuclear fuel waste in an engineered vau lt at a depth of 500 to 1000 m in the plutonic rock of the Canadian Sh ield. In common with engineered barrier system designs being developed in other countries, the waste would be contained within durable conta iners that, in turn, would be isolated from the host rock by clay-base d materials. The objective of the CNFWMP is to develop a disposal conc ept that will protect human health and the natural environment far int o the future. Assessments of the conceptual vault designs are based on system theory in which an attempt is made to correlate experiences wi th theoretical concepts of planned systems in such a way that the resu lting coordination is sound and convincing. By necessity, since experi ments with a total disposal system can never be performed, both the de sign and the performance assessment rely on experiments performed on p hysical models of vault elements over relatively short times and on in formation inferred from calculations (mathematical models) that simula te the probable behaviour of the system in the space-time domain of in terest. For a simulation model to be successful, that is applied withi n a real world situation, the model must provide information regarding the behaviour of the system of interest that is clearly better, in so me way, than the mental image or other abstract model that would be us ed instead. The results of a series of tests performed within the acti vity known as validation serve as tangible evidence regarding the succ ess of a model in representing the system of interest. This paper focu sses on the validation of the models that describe the hygro-thermo-me chanical behaviour of the engineered clay-barriers proposed for applic ation in the Canadian disposal system concept. The strategy being used to address the key issues in modelling to minimize the model error an d to maximize the usefulness of the simulation model, based on testing procedures, is reviewed. Finally, a concept of the validation of code s/models that describe the unsaturated behaviour of engineered clay ba rriers, is described.