Experimental-artificial intelligence approach for characterizing electrical resistivity of partially saturated clay liners

The aim of this study is to investigate the evolution of electrical resistivity of different kaolinite-dominant clay liners, in terms of its soil composition, as its moisture content and dry density change. Eight different mixtures of Kaolin-Bentonite (90%K-10%B; 80%K-20%B; 70%K-30%B; 60%K-40%B), and Kaolin-Sand (90%K-10%S; 80%K-20%S; 70%K-30%S; 60%K-40%S) were tested in this study. Artificial Neural Network, ANN, method was used to develop an electrical resistivity model using the experimental results. The developed model offers the required level of generalization to analyse and assess precisely the effects of different variables on the electrical resistivity of kaolinite-dominant clay liners. The outcomes of this study highlight the effects of water content, soil composition, and dry density on the electrical resistivity of soils. The results in this study show that, at low water content, the adsorbed water and interparticle contacts provide continuous pathways for electrical flow through the soil. Furthermore, the results also indicate that increasing the bentonite content in the mixture decreases its electrical resistivity whereas increasing the sand content increases its electrical resistivity. This behaviour could be attributed to the highest surface conduction of bentonite clay compared to sand. For the soil mixtures tested in this study, the results also show that increasing the dry density of the soil by 20% could result in 50% reduction in its electrical resistivity. This behaviour could be explained in terms of the expected increase in number and area of interparticle contacts as dry density increases that could also improve soil pore water connectivity.