Prediction of unconfined compressive strength and California bearing capacity of cement- or lime-pozzolan-stabilised soil admixed with crushed stone waste

摘要

ABSTRACT So far, a limited number of studies have been published on utilisation of crushed stone waste, cement, lime and pozzolan as stabilisers for soil stabilisation. The research conducts soil engineering properties and strength test for various contents of crushed stone waste, cement, lime and pozzolan under various curing times. Increasing the pozzolan or cement content resulted in an increase in UCS; however, the UCS increased as the lime content increased from 0% to 5% and then decreased as lime content violated from 5%. An increase in the granite crushed stone content from 0% to 10% resulted in an increase in the UCS value. The unsoaked CBR and UCS values are almost 1.37 and 1.24 times more than the corresponding soaked ones for all cases, respectively. The CBR value increased with the increase in the pozzolan, lime or cement content. The cement content has more important influence on the increase in CBR value than the lime or pozzolan content. The multi-layer perceptron (MLP) neural network and non-linear regression (NLR) techniques are employed to develop models to predict the CBR and UCS values of the stabilised specimens.