Anthony T. C. Goh

Back-propagation neural networks for modeling complex systems

Abstract In complex engineering systems, empirical relationships are often employed to estimate design parameters and engineering properties. A complex domain is characterized by a number of interacting factors and their relationships are, in general, not precisely known. In addition, the data associated with these parameters are usually incomplete or erroneous (noisy). The development of these empirical relationships is a formidable task requiring sophisticated modeling techniques as well as human intuition and experience. This paper demonstrates the use of back-propagation neural networks to alleviate this problem. Backpropagation neural networks are a product of artificial intelligence research. First, an overview of the neural network methodology is presented. This is followed by some practical guidelines for implementing back-propagation neural networks. Two examples are then presented to demonstrate the potential of this approach for capturing nonlinear interactions between variables in complex engineering systems.

Prediction of Ultimate Shear Strength of Deep Beams Using Neural Networks

This study investigates the feasibility of using neural networks to evaluate the ultimate strength of deep reinforced concrete beams in shear. A neural network is an information processing system whose architecture essentially mimics the biological system of the brain. The neural network is particularly useful for evaluating systems with a multitude of nonlinear variables as in this study, where the critical factors include the strength of the concrete, the beam geometry, and the steel reinforcement in the beam. No predefined mathematical relationship between the variables in assumed. Instead, the neural network learns by example patterns obtained from published experimental data of concrete beams tested to failure. Details of the neural network predictions were more reliable than predictions using other conventional methods

Search for critical slip circle using genetic algorithms

Abstract In using traditional nonlinear optimization techniques for determining the critical slip surface in slope-stability analysis there is generally some uncertainty as to robustness of the algorithms to locate the global minimum factor of safety rather than the local minimum factor of safety for complicated and non-homogeneous geological subsoil conditions. This paper describes the incorporation of a genetic algorithm methodology which is becoming increasingly popular in engineering optimization problems as it has been shown in a wide variety of problems to be suitably robust for the search not to become trapped in local optima. First, the general principles of genetic algorithms are described. The genetic algorithm procedure used to locate the critical circular slip surface is then described. The five examples presented indicate that the search strategy was suf ficiently robust and efficient to handle multi-layered soils.

Assessment of basal stability for braced excavation systems using the finite element method

Abstract Conventional methods of predicting the basal stability of braced excavations are unable to take into consideration the stiffness of the retaining wall and the depth of penetration of the wall below the bottom of the excavation. A simple and improved procedure for predicting the stability of strutted excavations using the finite element method is presented. Detailed studies were carried out to assess the effects of the wall properties and soil geometry on the stability of the excavation.

SOIL LABORATORY DATA INTERPRETATION USING GENERALIZED REGRESSION NEURAL NETWORK

Abstract Artificial intelligence techniques which incorporate empirical knowledge and/or pattern matching techniques are ideally suited to assist engineers to interpret information from site and laboratory investigations because of the “imprecise” nature of soil. This paper explores the pattern matching and prediction capabilities of neural networks to interpret laboratory test data. The neural network paradigm used in this paper is the generalized regression neural network (GRNN) algorithm. Detailed examples are given of the use of this approach to assist engineers to interpret laboratory test data from consolidation tests and to characterize soil types from laboratory particle size distribution information. The main advantage of the GRNN technique in comparison to the widely used back-propagation neural network algorithm is the speed at which the optimal neural network configuration is determined, since this process only involves adjusting one variable.

Determination of earth pressure balance tunnel-related maximum surface settlement: a multivariate adaptive regression splines approach

A major consideration in urban tunnel design is to estimate the ground movements and surface settlements associated with the tunnelling operations. Excessive ground movements may result in damage to adjacent buildings and utilities. Numerous empirical and analytical solutions have been proposed to relate the shield tunnel characteristics and surface/subsurface deformation. Numerical analyses, either 2D or 3D, have also been applied to such tunnelling problems. However, substantially fewer approaches have been developed for earth pressure balance (EPB) tunnelling. Based on instrumented data on ground deformation and shield operation from three separate EPB tunnelling projects in Singapore, this paper utilizes a multivariate adaptive regression splines (MARS) approach to establish relationships between the maximum surface settlement and the major influencing factors, including the operation parameters, the cover depth and the ground conditions. Since the method has the ability to map input to output patterns, MARS enables one to map all influencing parameters to surface settlements. The main advantages of MARS over other soft computing techniques such as ANN, RVM, SVM and GP are its capacity to produce a simple, easy-to-interpret model, its ability to estimate the contributions of the input variables, and its computational efficiency.

Multivariate Adaptive Regression Splines Application for Multivariate Geotechnical Problems with Big Data

Despite the rapid increases in processing speed and memory of low-cost computers, the enormous computational costs of running complicated numerical analyses such as finite element simulations makes it impractical to rely exclusively on simulation for the purpose of design optimization since many geotechnical problems are highly nonlinear and multivariate. To reduce the cost, surrogate models, also known as meta-models, are constructed and then used in place of the actual numerical simulation models. To ensure the surrogate model is more reliable, the ranges of the design variables should be as wide as possible. Thus meta-modeling techniques capable of analyzing multivariate problems are desirable. This paper explores the use of a fairly simple nonparametric regression procedure known as multivariate adaptive regression splines (MARS) in approximating the relationship between the inputs and outputs with a big data. First the basis of the MARS methodology and its associated procedures are explained in detail. Then two complicated geotechnical problems are presented to demonstrate the function approximating capabilities of MARS and its efficiency in dealing with multivariate problems involving large amounts of data. This paper demonstrates that the MARS algorithm is capable of producing simple, accurate and easy-to-interpret models and estimating the contributions of the input variables.

Advisory expert system for flexible pavement design

Abstract The adoption of an expert system approach to flexible pavement design could be especially useful, because of the complexity of the design algorithm involving numerous possible combinations of pavement material types and traffic data, and the heavy reliance on empirical correlations, which can vary from one region or state to another. This paper describes the development of an advisory expert system for flexible pavement design that attempts to mimic the design process of pavement engineering specialists. The highly interactive microcomputer expert system was developed, using an expert system shell.

Reliability assessment of ultimate limit state of twin caverns

ABSTRACT Construction of a cavern in close proximity to an existing cavern modifies the state of stresses and movements in a zone around the existing cavern, as some degree of interaction between these two caverns generally takes place. This study investigates the interaction of two parallel caverns and the influence of such interaction on stress-induced global stability in terms of a global factor of safety. A series of finite difference analyses were performed to derive the global factor of safety of a system of two parallel and adjacent caverns. A mathematical response surface model was then built using the multivariate adaptive regression splines (MARS) approach and a series of charts based on this surrogate model were developed to relate the global factor of safety to the critical parameters. The built MARS model is of high accuracy and is simple to interpret and can be used to perform probabilistic assessment of ultimate limit state of twin caverns.

Reliability Assessment of Diaphragm Wall Deflections in Soft Clays

For excavations in built-up areas with deep deposits of soft clays, it is essential to control ground movements to minimize damage to adjacent structures and facilities. This is commonly carried out by controlling the deflections of the retaining wall system. The limiting wall deflection or serviceability limit state is typically taken to be a percentage of the excavation height. In this study, extensive plane strain finite element analyses have been carried out to examine the excavation-induced wall deflections for a deep deposit of soft clay supported by diaphragm walls and bracing. Based on the numerical results, two polynomial regression approaches were used to develop the equations for estimating the maximum wall deflection. This paper describes how the developed equations can be used to perform reliability analysis of the diaphragm wall serviceability limit state to estimate the probability of exceeding the limiting wall deflection.