INTERNATIONAL JOURNAL OF INDUSTRIAL ENGINEERING AND PRODUCTION MANAGEMENT (IJIE) (INTERNATIONAL JOURNAL OF ENGINEERING SCIENCE) (PERSIAN)
Year:2007 | Volume:18 | Issue:2|Papers Count:12

Many meshless methods are presented in recent years for solving partical differential equations. All of them use a background mesh to introducing the Guossian points for solution of the integral equation. Numerical integration has significant effects on convergence and accuracy of solution from many meshless methods. They are meshless only from the point of view of interpolation of the field variables. In this paper we present a fully meshless procedure for linear and nonlinear convection-diffusion equation problems in steady and transient forms. In this new method a fully Least squares method is used in both function approximation and the discretization of the governing differential equations. The discritized equations are obtained via a discrete least squares method in which the sum of the squared residuals are minimized with respect to unknown nodal parameters in the inner and boundary nodes. In this process no numerical integration is needed and the obtained equations are symmetric and positive definite. The shape functions are derived using the moving least squares (MLS) method of function approximation with a spline weighting function. The convergency and accuracy of the method is programmed using visual Fortran 6.5and accuracy of the results are compared and verified with reference to some examples.

Utilizing artificial neural networks (ANN) simplify data processing operations in a broad range area in engineering. ANN is an oversimplified simulation of the human brain and composed of simple processing units referred to as neurons.This method was successfully applied to develop prediction of the patterns, functions approximation, optimization, future value prediction, automotive control and information recovery. Un-drained cohesion value plays an important role in geotechnical earthquake design of earth fill structures.In addition to laboratory test methods, in-situ testing methods as CPT and SPT provide reliable results. Different equations, charts and methods were provided by many researchers to describe the contingency between in situ test results and geotechnical parameters in recent decades. The present study investigates the relation between overburden pressure (s°), number of SPT (NSPT), plasticity index (PI), liquid limit (LL) and moisture content with un-rained cohesion value of clay (Cu). On the basis of VU triaxial test and unconfined compression test results presented in geotechnical studies from different sites of Iran, an artificial neural network model was developed to comprise the laboratory test results and the results provide by the model. The artificial neural network model results presented in this study exhibit a good agreement with those of laboratory test.

This paper considers the quadrilateral membrane elements with drilling degree of freedoms in the viewpoint of algebraic patterns. The presence of the free parameters in components of the stiffness matrix unifies the methods and provides the opportunity to improve the element performance. The element stiffness matrix decomposes into basic and higher order parts. The basic stiffness matrix satisfies the convergence requirements and has one free parameter. The higher other stiffness matrix, constructed by Assumed Natural Deviatoric Strain approach with 24 parameters, satisfies the accuracy and stability requirements. The element stiffness matrix has not any spurious modes and does not lock in the analysis. This template can be correlated with another methods and obtain the signature of the elements.

Since reversible water resource are limited and mostly utilized, to provide the increasing population of the world with enough water and food, it is necessary to apply advanced and innovative methods to produce most from the least water and decrease the wastewater due to the customary methods in design, execution, and operation of these resources.Because the lack of means for fitting operating rules and regulation with time variant water needs, a considerable portion of available water is wasted in th irrigation networks. That is why research on methods of operation and determination of types of controls and optimization approaches become necessary. In this research, while criteria for control systems design is introduced, a new control algorithm for irrigation system management is developed and used in one of the largest conveyance systems in the world. Utilizing simulation results, several different operation alternatives are investigated by SOBEK hydrodynamic model and performance assessment indices.The type of central control system in this research is control from downstream with combined feedback and feed forward (PI) with decoupling.The obtained results from this algorithm in attenuation of infra structural disturbances and hydraulic turbulences in studied canal come up to be of a high potential with considerable precision and accuracy. Use of this algorithm makes demand oriented water distribution and better performance of the system possible.

For an efficient design of space structures, the assessment of the displacements is an important issue. The use of standard methods is usually time consuming and limits the ability of the designer. Therefore it is ideal to have efficient methods for this purpose.In this paper, efficient neural networks are trained for predicting the displacements of domes under the action of wind loading. For these purpose two neural networks known as Back propagation and radial basis functions are trained and tested. The performance of these networks are assessed and compared for predicting the displacements of dome structures.

Strength and ductility enhancement and the plastic behavior of concrete in RC members are the consequences of concrete confinement. In the current study, the prediction of the behavior of concrete confined with CFRP composites has been evaluated using nonlinear analysis. To do so, the results of 29 experimental studies on the concrete confined with unidirectional CFRP were selected from the literature. The stress-strain relation of the selected cases was approximated with bi-linear curves; accordingly some' equations were extracted for nonlinear behavior of confined concrete by regression. The credibility of the proposed equations was verified using 16 other independent experimental data.The proposed equations are capable to be used in the concrete confinement modeling, using Tsai-Wu failure criterion for the hoop strain fracture of the bonded CFRP to concrete. To fulfill the current study, some concrete specimens confined with FRP composites with particular characteristics were modeled in ANSYS software.The results showed the suitability of the model, therefore the extracted stress-strain curves from the software could be fairly used in the parametric studies on concrete confinement under axial load and flexural moment.

The ANN methodology, inspired by neurobiology theories of massive interconnection and parallelism has been successfully employed in variety of optimization problems. In ground water management models, either governing equations are embedded into the management model or unit response matrixes are employed. Unit response matrixes development requires huge amount of data and/or simulation runs. In this paper, ANN is employed to develop unit response matrix coefficients to be later used in the management model.To do it, a ANN model has been trained to predict the outcome of the flow code, which results in unit response matrix coefficients for the aquifer under consideration. To train the ANN model different realizations from pumping well co-ordinates, distance between pumping and observation wells, and hydraulic conductivities of pumping wells were used, it was concluded that pumping well co-ordinates may be successfully employed for developing unit response matrix coefficients to be later used in management models.To test the performance of the proposed approach, the hypothetical aquifer was assumed. The aquifer response to different pumping stresses were compared using a well-defined simulation model and those resulted from unit response matrixes developed by (1) ANN approach and (2) direct data from groundwater simulation runs. It was concluded that ANN may be successfully employed for development of unit response matrix with limited data from field study or ground water simulation runs.

Due to special behavior of hot-mix asphalt (visco-elastic property), road body's reaction under load is a function of loading time duration. On the other hand equivalent load factors (ELF) depend on pavement distress and pavement reaction. Therefore this research investigates loading time duration effects on ELF. Since load time duration varies with vehicle speed therefore this thesis discuss about two aspects of load time.1. Dynamic load effects on ELF.2. Static loading time duration on ELF.Due to lack of required equipments and resources for practical research in this area in Iran, theoretical method was used in this research. A piece of AASHTO road was selected for comparing AASHTO ELF with estimated ELF in this thesis. After calculating pavement's critical reaction, in the speed range of 20 km/h to 120 km/h, ELF was determined. In this research, for evaluating the above mentioned reaction, KENLAYER software was employed.On the basis of theoretical method, final results clarify that loading time duration effects on equivalency load factors method is different at various criteria and axle.

SPARs (or column-type offshore platforms) are a kind of common platforms in the Oil and Gas exploration industries. Nowadays due to financial and technical reasons, these platforms are one of the most usual offshore platforms, so the analysis methods of their design are rapidly being developed, improved and modified. In this research, we have evaluated the motion behavior and the response of SPARs induced by waves by means of analytical capabilities of MOSES software. MOSES is a powerful computational tool for any system which can be placed in the fluid and can cope with variety of complex and time-consuming analyses (precisely and rapidly), such as Time Domain, Frequency Domain and Coupled analyses. In this research the concept of Dynamic Instability of SPARs which hadn't been able to be evaluated completely by common types of software, have been considered and estimated by MOSES software. As well, various developed geometric hull forms (compared to simple-classical hull forms) have been considered in order to obtain the improved and better motion responses (especially at Heave response). Both Time Domain and Frequency Domain analyses have been performed, and the consistency of results confirms the accuracy. With Coupled analysis, the values of responses have decreased due to the added damping produced by simultaneous modeling of platform's hull, Mooring and Risers systems.

Finding a method which is able to design and check the ability of filters in erosion control of base soil in embankment dams is one of the problems in geotechnical engineering. Laboratory tests and using criteria to design filters are two principal methods frequently used. For these two groups, NEF test and Sherard-Dunnigan criteria are of the most comprehensive methods. Although the advantages and disadvantages of such methods are not completely clear, for simplicity, designers prefer to use criteria more than laboratory tests. To design and choose proper filters during designing of some embankment dam projects in Iran, some NEF test were carried out by authors. The results however show some incompatibility with design criteria. This study attempts to assess the ability of each method. It seems that, NEF according to its fully and real modeling is more successful than criteria in prediction of filter behavior.

The subject of the presented paper is the seismic stability of reinforced earth slopes and walls using pseudo-static and limit equilibrium method. The main concept of the proposed method is to determine the most critical slip surface. The related sliding wedge is divided into a number of horizontal slices, which do not intersect the reinforcements. The force and moment equilibrium equations for each slice or the whole wedge depending on the formulation can be satisfied. Based on the number and nature of the equations and assumptions made, several formulations can be derived to determine the forces in the reinforcements. Solving the set of equilibrium equations, the total force in the reinforcements required to maintain the stability of the reinforced earth structure is calculated. The proposed formulations are evaluated using a comparison with published data of several famous papers. The comparative analysis shows a satisfying agreement with the published data.

The role of the level of reliability of meeting a desired hydropower system's firm energy yield and quantifying the uncertainty of stream-flows in design components of a hydro-electric project is assessed in this study. Sequential stream flow routing (SSR) is used in a reliability-based simulation (RBS) model by which the energy potential of a hydro-power system may be determined at different normal water levels. To investigate how important the adopted reliability level would be, the RBS model was executed at different long-run reliability levels. For the system under study, i.e. the Baktiari hydropower dam, one can realize that the system's installed capacity, firm and secondary energies will respectively change by 44, 38 and 10 percent if the reliability level varies between 80-100 percent. The problem was also explored by imposing an additional medium-term (annual) reliability constraint. It was found that the additional reliability constraint would change more the installed capacity and energy values generated. The higher the medium term reliability level, the more decrease in installed capacity and the system's firm energy yield. Subsequently, the impact of uncertainty in stream-flow process was investigated by generating a large number of synthetic correlated and non-correlated stream-flow series and then estimating the statistical moments of the design variables instead of the ones estimated based on a single historical time series of stream-flows. The results indicate the system's installed capacity and annual firm energy yield may vary by 14-17 percent in case of correlated stream-flows and by 8-17 percent in case of uncorrelated stream-flows. This is because of uncertainty in stream flow process represented by different samples synthetically generated.