Mechanical parameter identification technique for a bentonite buffer based on multi-objective optimization

For the safe disposal of high-level radioactive waste, it is necessary to establish a numerical model that can simulate the phenomenon of buffer expansion due to groundwater inflow. The Barcelona basic model (BBM), one of many models that describe the swelling behavior of buffer, can represent the behavior of expansive soil but requires various hydro-mechanical input parameters. Conventional experiments to determine these parameters are time-consuming and complicated; therefore, this study proposes a method of determining the BBM parameters by comparing the results of swelling tests and numerical analysis. The relationships between these parameters were determined through an artificial neural network and multi-objective optimization was used to derive the Pareto optimal sets. Among various optimized solutions, single BBM parameters were derived from Pareto sets by considering additional conditions. The results of the numerical analysis using the identified parameters and the experimental results exhibited similar trends.