In this paper, an extension of classical waste-load allocation models for river water quality management is presented to determine the monthly treatment or removal fraction of wastewater to evaporation ponds. The dimensionality of the problem, which is due to a large number of decision variables, is tackled by developing a new GA based optimization model, which is called a Sequential Dynamic Genetic Algorithm (SDGA). This is a deterministic multi-objective optimization model, which is linked to an unsteady water quality simulation model. The model minimizes the total losses incurred during the optimization time horizon, including the treatment or removal fraction costs and the costs associated with the deviation from water quality standards. The proposed model has been used for the water quality management and salinity reduction of the Karoon River in Iran. The results show the proposed model can effectively reduce the computational burden of the seasonal waste-load allocation problem. It is also shown that the seasonal waste-load allocation can significantly reduce the number and duration of standards violations.

One of the most powerful methods to implement the free surface is the Volume Of Fluid (VOF). In this study, an algorithm is developed, which includes an implicit pressure based method (SIMPLE) with a staggered grid and a Lagrangian propagation VOF method. Based on this algorithm, a computer code is generated and a cavity with a free surface and two test cases of dam-breaking problems are examined and, then, the effect of fluid sloshing on a near wall is also analyzed and a time history of the normal force on the wall is presented. The results show good agreement with experimental and other computational results."

This paper concerns the design of a neural state observer for nonlinear dynamic systems with noisy measurement channels and in the presence of small model errors. The proposed observer consists of three feedforward neural parts, two of which are MLP universal approximators, which are being trained off-line and the last one being a Linearly Parameterized Neural Network (LPNN), which is being updated on-line. The off-line trained parts are able to generate state estimations instantly and almost accurately, if there are not catastrophic errors in the mathematical model used. The contribution of the on-line adapting part is to compensate the remainder estimation error due to uncertain parameters and/or unmodeled dynamics. A time delay term is also added to compensate the arising differential effects in the observer. The proposed observer can learn the noise cancellation property by using noise corrupted data sets in the MLP's off-line training. Simulation results in two case studies show the high effectiveness of the proposed state observing method.

The objective of this research is to study the ability of a meshless method, called finite point method, in solving incompressible fluid flow problems using two stabilization schemes. The main goal of meshless methods is to reduce or remove the cost of grid generation. This issue is implemented using the satisfaction of governing differential equations on a regular or irregular set of nodes by interpolation functions, based on special least-squares approximations. In this research, the finite point method is used to solve the Stokes and the Navier-Stokes equations by employing two different stabilization schemes. In addition, the effects of least-squares approximations are studied.

A series of consolidated undrained triaxial tests were performed in order to understand the behavior of a gravely sand. The material was selected from Tehran alluvium and is classified as gravely sand in the Unified Soil Classification System. Critical state concepts were used for interpretation of the behavior of the soil. The critical state line in q-p' space was reasonably unique. However, it was not possible to define a unique critical state line in e-ln p' space. The overall scatter, in the critical state line for the gravely sand studied, was \pm 0.04, in terms of void ratio. Two reasons can be identified for this scatter. The first reason is the inevitable error in void ratio calculation and the second is that the behavior of this material, even at a critical state, is fabric and structure dependant. On the basis of these tests, it has been concluded that, for this material, a critical state zone with upper and lower limits can be defined. Nevertheless, a unique boundary surface can be defined by presenting the data in normalized stress space.

The present paper investigates, numerically, the effect of the Grashof number to a Reynolds number ratio (Gr/Re), on fluid flow and heat transfer within a vertical channel for two cases: Mixed (natural-forced) convection and combined mixed convection-radiation. The flow in the channel is assumed to be two-dimensional, laminar and steady. The wall temperature is defined as a linear function of the channel height. When dealing with the combined mixed convection-radiation case, radiational properties have been taken into account, both for the walls and the fluid. The fluid has a Prandtl number of 0.71 and it is radiationally assumed as a participating medium. A comparison between the two cases at a constant Gr/Re is reported, so as to investigate the influence of radiation, as one of the heat transfer modes, more clearly. To solve the governing equations (i.e., mass continuity, momentum and energy) the Finite Volume method is employed and the SIMPLE algorithm is adopted to couple the velocity and pressure fields. The radiative transfer equation is solved using the Discrete Ordinates Method, by adopting its S_4 order quadrature scheme. The results for both cases are presented as the profiles of axial velocity across the channel width, axial centerline velocity, bulk temperature and pressure versus channel height.

This method introduces the structural error of regression deviation, which is an effective method for the path generation of a vast type of planar and spatial mechanism. The proposed method avoids point-by-point comparison and requirement of timing and reflects the difference between the two curves very effectively in the objective function. By decreasing the number of the design variables, this method would help considerably in decreasing CPU time. The objective function that is based on regression error would converge to a global minimum by a genetic algorithm. At the end, the effectiveness of the method is shown by two numerical examples.

This paper presents an overview of a study on the design and analysis aspects of the Lake Urmia Bridge in Iran. For years there have been several detailed investigations on this subject. Here, these alternatives are discussed and, then, results of analyses for a proposed solution, a floating bridge, are presented. These aspects include environmental loads, structure and the mooring system.

In this paper, an explicit finite element based numerical procedure is presented for simulating three-dimensional inviscid compressible flow problems. The implementation of the first-order upwind method and a higher-order artificial dissipation technique on unstructured grids, using tetrahedral elements, is described. Both schemes use a multi-stage Runge-Kutta time-stepping method for time integration. The use of an edge-based data structure in the finite element formulation and its computational merits are also elaborated. Furthermore, the performance of the two schemes in solving a benchmark problem involving transonic flow about an ONERA M6 wing is compared and detailed solutions are presented.

The three-dimensional supersonic turbulent flows over wrap-around fin missiles have been computed using the Thin Layer Navier-Stokes (TLNS) equations to reduce the computational efforts compared to those of the Full Navier-Stokes (FNS) equations. In this research, the missile configuration is divided into multi regions to enable fluid flow simulation using Personal Computers (PC). It also makes it possible to use a different number of nodes and distribution of grids in each region to enhance the accuracy. The Thin Layer Navier-Stokes equations in the generalized coordinate system were solved using an efficient, implicit, finite-difference factored algorithm of the Beam and Warming. For the turbulent flow field computations, the well-known isotropic two-layer algebraic eddy viscosity Baldwin-Lomax model was used. The computations were performed for supersonic turbulent flows over wrap-around fin configurations for free stream Mach numbers, M_{\infty}=1.9-2.86. Predicted roll moment and longitudinal aerodynamic coefficients were compared with the experimental data at various angles of attack. The computational results are in good agreement with experimental data.

In this paper, crew work posture, as one of critical human factor considerations, will be reviewed in different ship spaces. Whether crew members are standing or seated at a workbench or machine, their working posture is extremely important. If the available hardware forces crew members to remain in an awkward position for a long period of time, they will obviously become fatigued and, thus, more apt to make mistakes or incur some type of physical disability. In such spaces, different work postures are one of the most important parameters in affecting crew efficiency and must be studied for each space. In this work, each working space, in some real ships, regarding different work postures has been studied and the profile of each workplace has been determined. In this work, by allocating a grade for each workplace, awkward spaces in ships may be determined.

Steady state boundary layer equations over a flat plate with a constant wall temperature can be solved by an integral solution (with three profiles for velocity and temperature), a similarity solution (exact) and a Blasius series solution. The analysis of entropy generation for each solution is carried out. The results show that the exact solution (similarity) is the one that minimizes the rate of total entropy generation in the boundary layer. Then, the Blasius solution has the least entropy generation of all. The bell-shaped profile (sinus profile) in the integral solution generates less entropy than the piecewise linear profile, consequently. So, with this method, if the exact solution for a specified problem were not available, one could evaluate the approximate solutions and recognize the best one among them. By introducing a new non-dimensional number (Ej number), which is the ratio of thermal entropy to friction entropy generation, one can recognize which of them is dominant in the boundary layer. Also, it is observed that variation of the total entropy generation is the same as the variation of boundary layer thickness, so, the non-dimensional total entropy generation for various solutions is constant.

In recent years, the use of larger Reinforced Concrete (R.C.) column-supported hyperboloid cooling towers has been increased significantly. Thus, the investigation on failure criteria for structural components of such structures under different loads has been found as an essential need. Construction of cooling towers in seismic zones initiated the study on the dynamic behavior of such structures due to seismic loads. In this paper, finite element analyses have been performed to obtain the stress concentration, nonlinear behavior, stability or safety factor of the R.C. tower due to earthquake loads. Outcomes of the study show that considerable plastic hinges were created in the X shape long columns of the R.C. hyperboloid cooling tower due to seismic loads, which resulted in a significant decrease in the stability safety factor and, thus, an increase in concerns.

The present work is addressed to the numerical study of transient laminar natural convection in an open space and induced by a line heat source. The governing equations, full Navier-Stokes and energy equations with primitive variables, are discretized in a staggered grid by a control volume approach. The equations for the fluid and solid (line heat source) phases are solved simultaneously using a segregated technique. Some of the physical and thermo-physical properties of the fluid (air), such as density, thermal conductivity and viscosity, were considered to vary with temperature. The results show that the energy equation reaches the steady state condition more rapidly than the momentum equations. Hence, at that time, the distribution of temperature does not show any change within the accuracy of the solution, while the distribution of the velocity still varies. The steady-state results obtained via the time-marching solution show good agreement with the published steady-state, self-similar results in the vicinity of the centerline of the plume. Also, the steady-state streamlines compare well with the published experimental results.