Determining the Variability of the As-Placed Dry Density of Gap Fill Material
Abstract
Canada is one of the leading countries in nuclear energy usage for the past 60 years. Intermediate and high levels nuclear waste need to be contained and isolated in a deep geological repository, DGR, (i.e., 500 to 800 m below the ground level), using a multiple-barrier isolation system. Gap Fill Material (GFM) is a Bentonite based material that will be used to backfill the remaining spaces between the central bentonite blocks containing fuel waste containers and surrounding excavation walls. The Nuclear Waste Management Organization (NWMO) requires a simple and easy method to remotely determine the density of the GFM inside the DGR to ensure that it satisfies the density requirements, and consequently, limits the radiation of buried nuclear fuel into the surrounding environment. Several direct and indirect density measurement techniques were investigated, and a comprehensive laboratory testing program was performed. The program includes 277 Shear wave velocity and 59 Cone Penetration Tests on GFM samples with densities between 1.45 and 1.8 g/cm3. The results indicate strong relationships and provide some correlations between the GFM density, shear wave velocity, and cone penetration resistance. The laboratory results were then used to develop 5 artificial neural networks employing regression tool in MATLAB software with a high correlation coefficient (R-value) between the measured and predicted output variables.