Evaluation of multivariable regression in predicting rock slake durability index

Abstract

The durability of stone against erosion, weathering and dissolution against water flow is an important parameter in Civil engineering projects. It is absolutely necessary to check the quality of slake durability rock against water flow in all types of stone breakwaters, rips - rap in dams, including building materials (a wide range of decorative stones). Slake durability testing is a common method for assessing the durability of rocks. The durability of stone has a lot to do with the texture, weathering and physical and mechanical properties of the stone. In this study, by performing slake durability tests on samples belonging to the Tale Zang Formation, the slake durability index (SDI) of limestone belonging to this formation was investigated. The effect of the number of cycles and the relationship between slake durability index and physical and mechanical properties have also been investigated. Univariable and multivariable regression relationships were used to predict the slake durability index. The results show that multivariable linear regression using a combination of geomechanical properties offers a higher correlation between these properties and the slake durability index (SDI) of the studied limestones. In multivariable regression models, n, UCS, E and BTS properties in different relationships have the greatest effect on the SDI value