Numerical Prediction of the Large-Scale in Situ PRACLAY Heater Test in the Boom Clay

The large-scale PRACLAY heater test has been running since November 2014 in the HADES URF (Belgium) to examine the combined impact on the host Boom Clay of the thermo-hydro-mechanical (THM) perturbations due to gallery construction and a large-scale thermal load which simulates the heat dissipation of radioactive waste. To get an insight into the expected evolution of the test, which will be helpful to guide the test, numerical prediction of the PRACLAY heater test had been performed before the heater was switched on. The numerical prediction, based on the existing related knowledge before the switch-on of the heater test, took into account the complex coupled THM processes and interactions in all the involved materials of the experimental set-up (Boom Clay, bentonite, concrete lining, backfill sand, etc.) and the test phases from the gallery excavation till the end of heating which is planned in 2025. A large number of constitutive models and associated parameters required to define the numerical problem had been identified and calibrated on the basis of literature review, results of extensive laboratory tests and back analysis of many relevant in situ measurements from the HADES URF. The results of the numerical prediction are constantly and extensively compared with the in situ measured data obtained since the switch-on of the heater till 4-year stationary heating. Overall, it can be stated that the general Boom Clay THM responses in the PRACLAY heater test are in good agreement with the numerical prediction, which gives good indications on the model capacity and confirms the existing knowledge of the Boom Clay THM behaviour.

Automated summary

The PRACLAY heater test in HADES URF in Belgium has been running since November 2014 to examine the THM impact on Boom Clay due to gallery construction and thermal load simulating radioactive waste heat dissipation. Numerical predictions were done before the test began, based on existing knowledge and calibrated with laboratory tests and in situ measurements. Results show good agreement between the Boom Clay THM responses in the test and the numerical predictions, confirming the existing understanding of Boom Clay behavior.