Anuar Kasa

Tenfold cross validation artificial neural network modeling of the settlement behavior of a stone column under a highway embankment

Construction of embankments in engineering structures on soft clay soils normally encounters problems related to excessive settlement issues. The conventional methods are inadequate to analyze and predict the surface settlement when the necessary parameters are difficult to determine in the field and in the laboratory. In this study, artificial neural network systems (ANNs) were used to predict settlement under embankment load using soft soil properties together with various geometric parameters as input for each stone column (SC) arrangement and embankment condition. Data from a highway project called Lebuhraya Pantai Timur2 in Terengganu, Malaysia, were investigated. The FEM package of Plaxis v8 program analysis was utilized. The actual angle of internal friction, spacing between SC, diameter of SC, length of SC, and height of embankment were used as the input parameters, and the settlement was used as the main output. Non cross validation (NCV) and tenfold cross validation (TFCV) were used to build the ANN model. The results of the TFCV model were more accurate than those of the NCV model. Comparisons made with the predictions of the Priebe model showed that the proposed TFCV model could provide better predictions than conventional methods.

Prediction of External Stability for Geogrid-Reinforced Segmental Walls

Prediction of external stability for segmental retaining walls reinforced with geogrid and backfilled with residual soil was carried out using statistical methods and artificial neural networks (ANN). Prediction was based on data obtained from 234 segmental retaining wall designs using procedures developed by the National Concrete Masonry Association (NCMA). The study showed that prediction made using ANN was generally more accurate to the target compared with statistical methods using mathematical models of linear, pure quadratic, full quadratic and interactions.

Preparation of residual soil samples by using modified method

The purpose of this study was to establish an alternative method in preparing the residual soil samples for laboratory tests. The soil was compacted by using the modified method in order to get the desired values of dry unit weight that was equivalent to the values obtained from standard compaction method. The advantage of using this method was that, due to its larger size, more samples could be taken and tested under the same compaction condition and the mould could be directly mounted on the shaking table with the addition of the air bags to avoid the occurrence of boundary effects. The soil was compacted in a 300mm x 300mm x 300mm mould by using the vibrating hammer. The results of this study showed that the average dry unit weight value of 13.4 kN/m3, 13.6 kN/m3 and 13.7 kN/m3 obtained from 5 rounds/layer, 6 rounds/layer and 7 rounds/layer of compaction of modified method, was equivalent to about 92%, 93 % and 94% of maximum dry unit weight obtained from standard compaction method, respectively.

Prediction of Internal Stability for Geogrid-Reinforced Segmental Walls

Prediction of internal stability for segmental retaining walls reinforced with geogrid and backfilled with residual soil was carried out using statistical methods and artificial neural networks (ANN). Prediction was based on data obtained from 234 segmental retaining wall designs using procedures developed by the National Concrete Masonry Association (NCMA). The study showed that prediction made using ANN was generally more accurate to the target compared with statistical methods using mathematical models of linear, pure quadratic, full quadratic and interactions.

Numerical Analysis on the Effect of Geometry Parameter on the Behaviour of the T-Shaped Cantilever Retaining Wall

The design of T-shaped cantilever retaining wall and support system requires careful analysis especially the geometry of the wall. The design starts by proportioning the wall dimension for which stability is checked for. Commonly, limit equilibrium method (LEM) is used to analyse the performance, however, the LEM analysis unable to predict the displacement of the wall. Thus, the numerical analysis using finite element method (FEM) incorporated in Plaxis software computer program was adopted to analyse the behaviour of the T-shaped cantilever retaining wall. This chapter describes the performance of a cantilever wall with different geometry parameter to predict the displacement pattern of the wall and the distribution of lateral earth pressure. Plain strain condition with 15 nodded triangular elements have been chosen in modelling the wall using Plaxis 2D. The result of the analysis indicates the lateral earth pressure distribution depended on the height of the wall. While the displacement of the retaining wall system showed the reasonable agreement with the recommended design of the geometry ratio.

Comparison between unit cell and plane strain models of stone column ground improvement

This article presents a comparative study of the behaviour of clayey soil reinforcements using stone column ground improvement by means of numerical analyses. Two-dimensional finite element analyses with commercially available software, PLAXIS, were performed on end-bearing stone columns using 15-noded triangular elements to investigate the impact of the modelling type on the stress concentration ratio and failure mechanism of an improved foundation system. Consolidation analyses were conducted throughout the study using Mohr-Coulomb’s criterion. The computed values of the stress concentration ratios were compared for different key parameters, including the diameters of stone columns, c/c spacing of columns, friction angle of stone column material, and undrained cohesion of soft soil. The major conclusions of this study were that the stone column in the unit cell model shared between 2.5 to 3.14 times more loads than the surrounding soil, whilst in the plane strain model it shared between 1.7 to 2.9 times more loads. The use of plane strain approach to model the stone column gave a more comprehensive representation of the stress distribution and load transfer between the soil and columns, in addition to being a better method than the unit cell concept to evaluate the failure mode in this system.

Simulation of Stone Column Ground Improvement (Comparison between Axisymmetric and Plane Strain)

Most of the numerical studies on stone columns are based on the unit cell concept. However, the impact of interactions between adjacent columns and between the columns and the surrounding soil has not been investigated thoroughly. In this study, the finite element software, PLAXIS-2D-V8.2, was used to simulate a stone column as a unit cell and as a plane strain model in order to specify the difference between the performances of each model. The key factors that were investigated included the diameter and c/c spacing of the stone columns, friction angle of the stone column material and undrained cohesion of the soft soil. The emphasis of this parametric study was on the settlement improvement factor and excess pore water pressure, since these are critical to the design of stone columns. The main findings of this study were that in the plane strain model, the settlement improvement factor ranged between 2.2 and 3.2, which means that the settlement was improved more than twice. Meanwhile, in the unit cell concept, the settlement improvement factor did not exceed 1.53. The results of the settlement improvement were compared with the theoretical solutions that are commonly used for studies into the behaviour of stone columns. The unit cell model showed a lower peak value of excess pore water pressure than the plane strain model.

Application of fuzzy set theory to evaluate the stability of slopes

An artificial intelligence tools, Adaptive Neuro Fuzzy Inference System (ANFIS), was used in this study to predict the stability of slopes. Data used in this study were 300 various designs of slope. Those designs were created by using Slope/W which calculated factors of safety using various limit equilibrium methods (LEM) such as Bishop, Spencer and Morgenstern-Price. The input parameters consisted of height of slope, H (1–10 m), unit weight of slope material, γ (15-22 kN/m3), angle of slope, θ (11.31°-78.69°), coefficient of cohesion, c (0-50 kN/m2) and internal angle of friction, (20°- 40°) and the output parameter is the factor of safety. To build the fuzzy inference system, 243 rules were used at 60 epochs. The number of membership function for the any input was three and the type of membership function for output was linear. ANFIS obtained regression square (R2) of one for Bishop, one for Janbu, one for Morgenstern-Price and one for Ordinary. The result proved that ANFIS may possibly predict the safety factor with good precision and nearly to the target data.

Prediction of Safety Factor for Slope Designed with Various Limit Equilibrium Methods

The purpose of this study is to predict the stability of slope using adaptive neuro fuzzy inference system (ANFIS). Based on limit equilibrium theory, four different methods of analyses, i.e. Morgenstern-Price, Janbu, Bishop and Ordinary were used to calculate the overall safety factor of various slope designs. Neuro-fuzzy inference system was used to map from a given input to an output. Important parameters such as height of slope (H), unit weight of soil (γ), angle of slope (θ), coefficient of cohesion (c) and internal angle of friction (ф) were used as the input parameters while overall safety factor was the output. ANFIS model to predict the stability of the slopes was generated from the calculated data. Results showed that factors of safety predicted using ANFIS agreed well with factors of safety calculated using Limit Equilibrium Methods (LEM).

Stability Analyses of an Earth Dam Using Limit Equilibrium and Finite Element Methods

The objective of this research is to study the relationships between the stability of earth dam and its soil strength parameters. The soil strength parameters include cohesion, unit weight of soil and angle of friction. GeoStudio, commercially available software, was used to obtain the overall factor of safety using limit equilibrium method (LEM) and finite element method (FEM). Tables to show the relationships among soil strength parameters for factor of safety 1.0 and 1.5 are presented in this paper.