Correlating soil properties with post-yield oedometer stress-strain behavior of Finnish sensitive clays through PoYOSS^Sen_Clay

Research Center Terra, Department of Civil Engineering, Faculty of Built Environment, Tampere University, Tampere, Finland

^* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

A novel post-yield oedometer stress-strain prediction model, PoYOSS^Sen_Clay, was developed at Tampere University, Finland, for Finnish sensitive clays. As a typical characteristic behavior of sensitive clays, in CRS oedometer tests, rapid deformation occurs immediately after yield stress. The post-yield phase can be approximated accurately by three parameters: a limit constrained modulus ML, a limit stress interval for destructration phase SL, and a modulus number M', proposed by Sällförs (1975). In one of the five principal submodels of PoYOSS^Sen_Clay, as the learning-based predictor, multi-layer perceptron (MLP) artificial neural network (ANN) was utilized to estimate ML, M', and SL. This study investigated the correlations between these three deformation parameters and various combinations of soil properties via the mentioned submodel. For this data-driven assessment, a database was employed comprising five Finnish testing sites of sensitive clays, with sensitivity ranging from 14 to 122. The analysis indicated which soil properties exhibited the strongest correlations with each of ML, M', and SL. In general terms, it was revealed that these three parameters were highly correlated with water content, wn, fallcone shear strength for intact state, Su,int, yield effective stress, σ’y, and plastic limit, wp.