Prediction of geotechnical properties of clayey soils stabilised with lime using artificial neural networks (ANNs). Issue 2 (4th March 2019)

• Record Type: Journal Article
• Authors: Taleb Bahmed, Ismehen; Harichane, Khelifa; Ghrici, Mohamed; Boukhatem, Bakhta; Rebouh, Redouane; Gadouri, Hamid
• Is Part Of: International journal of geotechnical engineering. Volume 13:Issue 2(2019)
• Page Start: 191
• Page End: 203
• Publication Date: 2019-03-04
• Subjects: Problematic soils -- lime -- artificial neural networks (ANNs) -- plasticity index (PI) -- maximum dry density (MDD) -- optimum moisture content (OMC); Geotechnical engineering -- Periodicals; Soil mechanics -- Periodicals; Rock mechanics -- Periodicals; 624.1513
• DOI: 10.1080/19386362.2017.1329966
• Languages: English
• ISSNs: 1938-6362

Abstract

Abstract: Clayey soils are known as problematic soils for geotechnical engineering since several years. The effect of mineral additives on geotechnical properties of clayey soils has been many times investigated. However, there are a few investigations about the use of artificial neural networks (ANNs) for predicting the geotechnical properties of stabilised soils, all the same, the ANNs can be successfully used in this field. The accurate prediction of plasticity index (PI), maximum dry density (MDD) and optimum moisture content (OMC) is beneficial for the construction engineering in order to avoid the cumbersome tests in the laboratory. The aim of this research is to develop three models with good performances based on ANNs, and to predict all the PI, OMC and MDD values of subgrade soil stabilised with the addition of lime, using basic soil parameters which are always available for engineers. Three different models are developed which each one corresponding to the best architecture for the three properties where these models can be used as a reliable tool to predict the PI, OMC and MDD of clayey soils stabilised with lime.