Scientia Iranica
Year:2009 | Volume:16 | Issue:1 (TRANSACTION A: CIVIL ENGINEERING)|Papers Count:11

In this paper, a numerical approach to the reliability analysis of prestressed reinforced concrete long span bridges is presented. A bridge is modeled by finite element software and the analysis is performed in time domain by considering the bridge material nonlinearity. The considered random variables are: Specific strength of concrete, yield stress of steel bars, yield stress of prestressed bars, all sectional dimensions, and structural damping ratio, effective depth of steel bars and the magnitude and PGA of earthquake. In this study, the reliability of a bridge structure is evaluated under earthquake excitations. For this purpose, the First-Order Second-Moment (FOSM) method is used. In this method, the mean value and standard deviation of the random variables are considered for evaluating structural reliability. The proposed procedure is applied to evaluate the reliability of an existing prestressed arch concrete bridge located in Bandar-e-Anzali in Iran. Bandar-e-Anzali is a very high-risk earthquake zone. The results of the study show that the structural damping ratio, magnitude and PGA of earthquakes have a significant effect on the variation of reliability in the structure, while variations in the dimensions of the structure have little effect on the reliability index.

Performance based seismic design usually requires nonlinear dynamic or static analyses to assess the performance level of the structure under seismic action. To trace the exact performance point of a structure, these analyses should sometimes be repeated several times over. Analysis iterations mainly depend on the initial design and performance of the structure. So, a method that can present an appropriate initial selection with minimum time and e_ort would be precious. Such a method would also be very e_ective for seismic structural assessment. In this paper, an intelligent system has been created for the estimation of plastic hinge distribution and lateral ductility distribution and, also, for the assessment of existing steel structures, based on a direct displacement based design procedure. The method has been applied to the steel braced frames with concentric bracing systems in low, medium and high rise buildings. The designer can use this knowledge based system to obtain the performance level of existing steel structures, according to proposed seismic code levels. Finally, the intelligent system has been veri_ed using nonlinear dynamic analysis.

There are several approaches for modeling the fatigue life and damage of asphalt pavements, such as stress-strain and damage mechanics. In this research, a simple mechanistic approach is used to explain the destruction of asphalt pavements. For asphalt reinforcement, two types of geosynthetic were used in the airfield at Imam Khomeini airport, Tehran. Non-reinforced, reinforced with a geogrid and geotextile specimens with dimensions of 50´63´381 mm were obtained from the asphalt slab field section. Fatigue tests of this study have been conducted with a four point beam test and a fatigue load with a half-sin wave at a frequency of 10 cycle/sec (no rest period), has been used. The results indicated that specimens reinforced with geosynthetics exhibit a higher initial stiffness module and lower crack propagation rate than non-reinforced specimens.

Artificial Neural Networks (ANNs) have recently been introduced as an efficient artificial intelligence modeling technique for applications involving a large number of variables, especially with highly nonlinear and complex interactions among input/output variables in a system without any prior knowledge about the nature of these interactions. Various types of ANN models are developed and used for different problems. In this paper, an arti_cial neural network of the feed-forward back-propagation type has been applied for the prediction of density and compressive strength properties of the cement paste portion of concrete mixtures. The mechanical properties of concrete are highly incensed by the density and compressive strength of concrete cement paste. Due to the complex non-linear effect of silica fume on concrete cement paste, the ANN model is used to predict density and compressive strength parameters. The density and compressive strength of concrete cement paste are affected by several parameters, viz, watercementitious materials ratio, silica fume unit contents, percentage of super-plasticizer, curing, cement type, etc. The 28-day compressive strength and Saturated Surface Dry (SSD) density values are considered as the aim of the prediction. A total of 600 specimens were selected. The system was trained and validated using 350 training pairs chosen randomly from the data set and tested using the remaining 250 pairs.Results indicate that the density and compressive strength of concrete cement paste can be predicted much more accurately using the ANN method compared to existing conventional methods, such as traditional regression analysis, statistical methods, etc.

In this paper, a new method is proposed for identification of inelastic shear frame structures with hesteresis, using data collected on their dynamic response. It uses the Prandtl-Ishlinskii rate independent model for hysteresis, which was originally used in the field of plasticity and ferromagnetism. The proposed identification method is capable of identifying the mass, damping and restoring force of a frame structure, which can be used in forming the equations of motion of the frame. By solving the equations of motion, the dynamic response is predicted. The method is based on the combined use of Quadratic Programming (QP) and Genetic Algorithms (GA). First, assuming a set of Prandtl-Ishlinskii constants, the QP is used to find the best frame parameters that can be used in its equations of motion to predict its dynamic response with the minimum of error compared to the real data collected on its dynamic response, while the GA is used to find the best Prandtl-Ishlinskii constants for more reduction in error. The method has been applied to different frames with bilinear nonlinearity where the results show the high capability of the method. Two examples, a Single and a Multi Degree Of Freedom (SDOF and MDOF) frame, are included in the paper.

This paper employs the Particle Swarm Optimization (PSO) method to solve the problem of the optimal design of cascade stilling basins. PSO is a relatively recent heuristic search method whose mechanism is inspired by the swarming or collaborative behavior of biological populations. The objective of this research is to minimize the total construction cost of cascade stilling basins, which is a function of height of the falls and length of stilling basins, while fulfilling the hydraulic and topographical criteria. To illustrate the application of PSO, a benchmark design is taken from the work of Vittal and Porey [1] on a cascade stilling basin for the Tehri Dam, India.

In this paper, the cyclic nonlinear behavior of a brace element has been modeled. A brace element is modeled as two elastic beam-column segments, which are connected to each other via a plastic hinge. The far ends of the element are hinged. By a suitable combination of the isotropic and kinematical hardening rules of plasticity, the nonlinear axial force-displacement relation for a beam element has been derived. So, the strain hardening, strain softening, tangential modulus of elasticity and Bauschinger effects are taken into account. This model shows good agreement with experimental results that have been reported in other research works.

A formulation of the Element Free Galerkin (EFG), one of the mesh-less methods, is developed for solving coupled problems and its validity for application to soil-water problems is examined through numerical analysis. The numerical approach is constructed to solve two governing partial differential equations of equilibrium and the continuity of pore water, simultaneously. Spatial variables in a weak form, the displacement increment and excess pore water pressure increment, are discretized using the same EFG shape functions. An incremental constrained Galerkin weak form is used to create the discrete system equations and a fully implicit scheme is used to create the discretization of the time domain. Implementation of essential boundary conditions is based on penalty method. Examples are studied and the obtained results are compared with closed-form or finite element method solutions to demonstrate the capability of the developed model. The results indicate that the EFG method is capable of handling coupled problems in saturated porous media and can predict well, both soil deformation and the variation of pore water pressure, over time.

Several series of measurements were conducted to explore the hydraulic characteristics of a V-shaped bottom channel by using low & high-speed velocity propellers for point-wise velocity measurements. Also, in order to understand the effect of cross sectional channel shape on the distribution of depth-averaged velocity in the experimental channel, cases with different fowl rates were examined. Using SURFER software, the contour plots of 2D isovels were drawn as interpolation among averaged depths and velocities, obtained from superposing the various profile sections. It was observed that isovels are parallel to the channel boundary in a region close to the bed, and almost symmetric about the centerline, with some deviations. The variation of point velocities in each slice considered along a span wise direction, in order to study the depth wise velocity profile distributions, is shown. The lateral variations of depth averaged velocities indicate that the velocity distributions are almost symmetrical about the cross sectional centerline, except for some flow cases, in which there are slight deviations, despite the fact that the flow condition was uniform for all cases. It was found that the widely used log-law for the vertical profile of velocity does not appropriately model the velocity distribution in this particular channel shape. Considering the results obtained for the span- and depth-wise velocity distributions, especially the distortion of the isovels and the location of maximum velocity, there are strong evidences of secondary currents that are present in this channel cross section.

Railway managers have a strong economic incentive to minimize track maintenance costs, while maintaining safety standards and providing adequate service levels to train operators. The objective of this study is to apply a procedure for making optimal maintenance decisions in Iranian Railways. This study consists of two parts. First, a cumulative damage model, based on a Markov process, is applied to model the deterioration of the track. For this reason, tracks are categorized into six classes, so that those tracks with similar traffic loads and geographical location are collected into one class. The track survey data from 215 blocks (4,228 km) of the ten divisions of the Iranian Railway system, during 2002-2004, is used to identify the transition matrix. Secondly, probabilistic dynamic programming is used to find the optimal repair for each possible track state in the planning horizon. This approach allows an optimal maintenance decision to be determined for the track at any point in time within the planning horizon.

The Marshall Stability of asphalt concrete is one of the most important parameters in mix design and quality control. This property depends on many factors such as gradation, percentage of crushed aggregates, asphalt content and construction quality. In this research, the variation of Marshall Stability with asphalt content is simulated using Artificial Neural Networks (ANNs) with a Levenberg- Marquardt Back Propagation (LMBP) training algorithm. The percentage of crushed aggregates; the percentage passing through sieve numbers 200, 50, 30, 8, 4 and 1/2 inch, and the percentage of asphalt content are considered as network inputs and Marshall Stability as the network output. In the first stage, the maximum generalization ability of each network with a specified number of neurons in the hidden layer is determined. Comparing these maximum values reveals that the network with 8 neurons in the hidden layer has the maximum generalization ability. In the second stage, the variation of Marshall Stability with asphalt content is simulated by applying a sensitivity analysis to the network with the maximum generalization ability. This simulation is in good agreement with theory.