Akbar A. Javadi

A hybrid intelligent genetic algorithm

Application of genetic algorithms to optimization of complex problems can lead to a substantial computational effort as a result of the repeated evaluation of the objective function(s) and the population-based nature of the search. This is often the case where the objective function evaluation is costly, for example, when the value is obtained following computationally expensive system simulations. Sometimes a substantially large number of generations might be required to find optimum value of the objective function. Furthermore, in some cases, genetic algorithm can face convergence problems. In this paper, a hybrid optimization algorithm is presented which is based on a combination of the neural network and the genetic algorithm. In the proposed algorithm, a back-propagation neural network is used to improve the convergence of the genetic algorithm in search for global optimum. The efficiency of the proposed computational methodology is illustrated by application to a number of test cases. The results show that, in the proposed hybrid method, the integration of the neural network in the genetic algorithm procedure can accelerate the convergence of the genetic algorithm significantly and improve the quality of solution.

A new genetic programming model for predicting settlement of shallow foundations

In this paper, a new genetic programming (GP) approach for predicting settlement of shallow foundations is presented. The GP model is developed and verified using a large database of standard penetration test (SPT) based case histories that involve measured settlements of shallow foundations. The results of the developed GP model are compared with those of a number of commonly used traditional methods and artificial neural network (ANN) based models. It is shown that the GP model is able to learn, with a very high accuracy, the complex relationship between foundation settlement and its contributing factors, and render this knowledge in the form of a function. The attained function can be used to generalize the learning and apply it to predict settlement of foundations for new cases not used in the development of the model. The advantages of the proposed GP model over the conventional and ANN based models are highlighted.

An Evolutionary-Based Data Mining Technique for Assessment of Civil Engineering Systems

Purpose – Analysis of many civil engineering phenomena is a complex problem due to the participation of a large number of factors involved. Traditional methods usually suffer from a lack of physical understanding. Furthermore, the simplifying assumptions that are usually made in the development of the traditional methods may, in some cases, lead to very large errors. The purpose of this paper is to present a new method, based on evolutionary polynomial regression (EPR) for capturing nonlinear interaction between various parameters of civil engineering systems.Design/methodology/approach – EPR is a data‐driven method based on evolutionary computing, aimed to search for polynomial structures representing a system. In this technique, a combination of the genetic algorithm and the least‐squares method is used to find feasible structures and the appropriate constants for those structures.Findings – Capabilities of the EPR methodology are illustrated by application to two complex practical civil engineering pro...

An evolutionary based approach for assessment of earthquake-induced soil liquefaction and lateral displacement

Prediction of liquefaction and the resulting lateral displacement is a complex engineering problem due to heterogeneous nature of soils and participation of a large number of factors involved. In this paper new models are developed, based on evolutionary polynomial regression (EPR), for assessment of liquefaction potential and lateral spreading. The models developed for liquefaction and lateral spreading are compared to those obtained from neural network and linear regression based techniques. It is shown that the developed models are able to learn the complex relationship between either of these problems and their contributing factors in the form of a function with high level of accuracy (mostly in excess of 90%). The results of the EPR model developed for the liquefaction determination are used to find a novel 3-D boundary surface that discriminates between the cases of occurrence and non-occurrence of liquefaction. The developed boundary surface is employed to calculate the factor of safety against liquefaction occurrence.

Identification of parameters for air permeability of shotcrete tunnel lining using a genetic algorithm

Abstract This paper describes an identification process for determination of material parameters in a constitutive relationship, describing time dependency of air permeability of shotcrete tunnel lining. A numerical model has been developed to predict the air losses from tunnel face and perimeter walls in compressed air tunnelling. Field data from a Tunnel in Germany has been used to verify and calibrate the numerical model. A relationship has been established to describe the variation of the air permeability of shotcrete tunnel lining with time and the technique of parameter identification has been used to determine the parameters of this relationship. A genetic algorithm has been used in the optimisation procedure. It has been shown that time dependency of permeability of shotcrete plays a key role in controlling the air losses in driving tunnels under compressed air with shotcrete as a temporary or permanent lining and this time dependency should be taken into account in design.

Modeling Groundwater Flow and Seawater Intrusion in the Coastal Aquifer of Wadi Ham, UAE

Groundwater pumping from Kalbha and Fujairah coastal aquifer of the United Arab Emirates (UAE) has increased significantly during the last two decades to meet the agriculture water demands. Due to the lack of natural replenishment from rainfall and the excessive pumping, groundwater levels have declined significantly causing an intrusion of seawater in the coastal aquifer of Wadi Ham. As a result, many pumping wells in the coastal zone have been terminated and a number of farms have been abandoned. In this paper, MODFLOW was used to simulate the groundwater flow and assess the seawater intrusion in the coastal aquifer of Wadi Ham. The model was calibrated against a five-year dataset of historical groundwater levels and validated against another eleven-year dataset. The effects of pumping on groundwater levels and seawater intrusion were investigated. Results showed that reducing the pumping from Khalbha well field will help to reduce the seawater intrusion into the southeastern part of the aquifer. Under the current groundwater pumping rates, the seawater will continue to migrate inland.

The Analytical Hierarchy Process for contaminated land management

Decision analysis (DA) methods and techniques are used to support decision-makers deal with complexities, uncertainties and risks of contaminated land management problems. Over the years, several methods have been used for environmental decision-making. This paper reviews the different methods and techniques used for contaminated land decision-making and decision analysis. We focus on the Analytic Hierarchy Process, which is among the most widely used and fastest growing decision-analytic techniques in several disciplines, including environmental and resource planning and management. However its application to contaminated land management problems as yet has been minimal and under explored. We explore the potential of this technique and explain it with a simple case study.

A new approach for prediction of the stability of soil and rock slopes

Purpose – Analysis of stability of slopes has been the subject of many research works in the past decades. Prediction of stability of slopes is of great importance in many civil engineering structures including earth dams, retaining walls and trenches. There are several parameters that contribute to the stability of slopes. This paper aims to present a new approach, based on evolutionary polynomial regression (EPR), for analysis of stability of soil and rock slopes.Design/methodology/approach – EPR is a data‐driven method based on evolutionary computing, aimed to search for polynomial structures representing a system. In this technique, a combination of the genetic algorithm and the least square method is used to find feasible structures and the appropriate constants for those structures.Findings – EPR models are developed and validated using results from sets of field data on the stability status of soil and rock slopes. The developed models are used to predict the factor of safety of slopes against fail...

Intelligent finite element method: An evolutionary approach to constitutive modeling

In this paper, a new approach is presented, for constitutive modeling of materials in finite element analysis, with potential applications in different engineering disciplines. The proposed approach provides a unified framework for modeling of complex materials, using evolutionary polynomial regression-based constitutive model (EPRCM), integrated in finite element analysis. Evolutionary polynomial regression (EPR) is a computing technique that generates a transparent and structured representation of the system being studied. The main advantage of EPRCM over conventional constitutive models is that it provides the optimum structure for the material constitutive model representation, as well as its parameters, directly from raw experimental (or field) data. The proposed algorithm provides a transparent relationship for the constitutive material model that can readily be incorporated in a finite element model (FEM). The incorporation of EPRCM into FEM will be presented and the application of the resulting methodology for material modeling in finite element analysis will be illustrated through two examples.

Multi-objective Optimization of Different Management Scenarios to Control Seawater Intrusion in Coastal Aquifers

Seawater intrusion (SWI) is a widespread environmental problem, particularly in arid and semi-arid coastal areas. Therefore, appropriate management strategies should be implemented in coastal aquifers to control SWI with acceptable limits of economic and environmental costs. This paper presents the results of an investigation on the efficiencies of different management scenarios for controlling saltwater intrusion using a simulation-optimization approach. A new methodology is proposed to control SWI in coastal aquifers. The proposed method is based on a combination of abstraction of saline water near shoreline, desalination of the abstracted water for domestic consumption and recharge of the aquifer by deep injection of the treated wastewater to ensure the sustainability of the aquifer. The efficiency of the proposed method is investigated in terms of water quality and capital and maintenance costs in comparison with other scenarios of groundwater management. A multi-objective genetic algorithm based evolutionary optimization model is integrated with the numerical simulation model to search for optimal solution of each scenario of SWI control. The main objective is to minimize both the total cost of management process and the total salinity in aquifer. The results indicate that the proposed method is efficient in controlling SWI as it offers the least cost and least salinity in the aquifer.