1. UKM Learning and Research Repository
2. PHD Thesis / Tesis PHD
3. Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina

Title:	Behaviour of stone columns supported embankment on soft soil
Authors:	Maryam M. Gaber Abdalwahab (P80324)
Supervisor:	Anuar Kasa, Ir. Dr.
Keywords:	Columns Embankments Soils Universiti Kebangsaan Malaysia -- Dissertations Dissertations, Academic -- Malaysia
Issue Date:	14-Jun-2019
Description:	The vibro-stone column SC technique is commonly adopted to enhance the bearing capacity and reduce settlement of soft soils. The technique has also been widely used to provide flexible economic and sustainable solutions applicable to various types of weak soils. However, previous researchers have focused on the techniques of SCs under shortterm conditions. Other researchers have adopted the bulging of SCs as the primary failure mode. More importantly, a solid theoretical basis for the ground reinforcement technique is still lacking. Therefore, this study has been undertaken to address this gap in knowledge. The first research objective is to analyse the different approaches required to simulate SC ground improvement. Additionally, the research aims to develop numerical models for the performance of individual and group of SCs supported embankment foundation by investigating the effect of varying parameters of soil and SC. Next, the validation of artificial intelligence methods ability to predict the SCs behaviour was studied. The data used was obtained from two embankment projects in Malaysia. In this work, the finite element method (FEM) and artificial intelligence (AI) systems are used to predict the parametric performance of SC reinforced soft soil foundations. Moreover, Plaxis 2D code was adapted to simulate the individual SC (unit cell) and a group of SCs (plane strain) with surrounding soft soil to study the behaviour of the ordinary SC under short and long-term conditions. The numerical results provided calculated stress concentration ratio (SCR), bearing capacity (BC), failure mechanism, safety factor (SF), and the settlement improvement factor (SIF) of an improved ground. The factors that influence the behaviour of SCs were investigated, including the spacing between adjacent columns, diameter, friction angle, unit weight, elastic modulus and Poissons ratio of the SCs and embankment fill height. Moreover, trial simulations were performed to predict the behaviour of SCs using the artificial intelligence (AI) technique, which includes artificial neural network (ANN) and response surface methodology (RSM). The main findings considered the column-column, and column-soil interaction, which significantly enhanced the strength of the embankment system. Hence, the unit cell concept is unsuitable for selected SC studies such as failure mechanism and stress distribution. Besides, sufficient information on the behaviour of a multi-stage construction procedure of an embankment conducted over a soft soil and strengthened by SCs was deduced. One significant finding of the numerical analysis is that the friction angle of the SC and the area replacement ratio are designated as the most critical design parameters. Also, the ANN and RSM methods achieved a high level of precision in modelling the behavioural parameters of SCs. In general, the prediction result for ANN was more accurate than RSM where the regression square (R2 ) results for ANN were 0.998, 0.995 and 0.891 for the SIF, BC and SF models, respectively. However, the R2 results for RSM were 0.939, 0.967 and 0.892, respectively. Lastly, the use of RSM helps to establish an approximate functional relationship, which significantly enhances the computational efforts.,Ph.D.
Pages:	223
Call Number:	TA660.C6A235 2019 3 tesis
Publisher:	UKM, Bangi
Appears in Collections:	Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.