Iranian Journal of Science and Technology Transactions of Civil Engineering
Volume:37 Issue: 1, 2013

In this paper optimum design of truss structures for both discrete and continuous variables based on the Cuckoo Search (CS) algorithm is presented. The CS is one of the recently developed population based algorithms inspired by the behavior of some cuckoo species together with the Lévy flight behavior of some birds and fruit flies. In order to demonstrate the effectiveness and robustness of the present method, minimum weight design of truss structures is performed and the results of the CS and the selected well-known meta-heuristic search algorithms are compared for both discrete and continuous design of three benchmark truss structures.

The formation of null basis for equilibrium matrix is the most important part of the finite element analysis when the force method is utilized. For an optimal analysis, the selected null basis matrices should be sparse and banded leading to sparse, banded and well-conditioned flexibility matrices. In this paper, an efficient algorithm is developed for the formation of null basis of triangular and rectangular plate bending finite element models, corresponding to highly sparse flexibility matrices. This is achieved by applying a modified ant colony system. An integer linear programming formulation is also presented to evaluate the quality of the results obtained by the proposed ant colony system algorithm. The efficiency of the present algorithm is illustrated through some examples.

Despite the large volume of work reported on the behaviour of reinforced concrete joints, only a few studies have been carried out to investigate the influence of retrofitting joints by FRP composites on the overall behaviour of an RC frame. To study the seismic performance of Moment Resisting RC frames retrofitted at joints by FRP, experimental and numerical investigations are carried out on a scaled-down frame of weak-beam, strong-column type, retrofitted by applying the FRP laminates at the web of the joints. Representing constitutive models are used to introduce the behaviour of concrete, steel and fibre-reinforced polymers in the numerical investigation. Full post-peak behaviour of the joints is captured considering strain softening of concrete. Finite element results show good agreement with the experimental findings. It is found that the maximum load carrying and displacement capacities of the joint after strengthening are increased and that the first steel yielding and development of crack occur at higher loads, further away from the joint, into the beam. The effects of different values of fracture energies on the behaviour of the reinforced concrete joint are also investigated. Nonlinear pushover analyses are also carried out to predict the seismic performance of an eight-storey and two additional low-rise frames retrofitted by steel braces and FRP. It is shown that by using FRP laminates at the web of the joints, the stiffness, the behaviour factor, R, performance level and the lateral load-carrying capacity of the damaged/plain frame are markedly increased.

In this paper, determination of the design variable optimum values of simple and spliced pre-tensioned girders in bridges is addressed by considering the life cycle cost of the whole structure. To this end, a program called OBPG is written with FORTRAN, consisting of the analysis, design and optimization subroutines based on the feasible direction method. AASHTO principles are used, assuming initial values for the design variables such as the dimensions of the girders and the deck, and the strands number. Then, using the analysis outcomes and considering the design criteria, the total cost of the structure, including that of the girders and slabs concreting, pre-tensioning cables performance, reinforcement and frameworks, are minimized as functions of the design variables. Although the initial optimization cost using life cycle cost is slightly more than that using the initial cost, optimization of the life cycle cost proves to be more economical. Ultimately, a program capable of analyzing simple and spliced pre-tensioned bridge girders is developed which calculates the life cycle and gives an optimum design.

Behavior of mortars with and without silica fume (SF), containing green ceramic powder (GCP) and marble dust (MD) under the conditions of pre-fire and post-fire was investigated. In addition, influence of re-curing on the mechanical properties and permeability of fire-damaged high-strength mortars was studied. For this purpose, mortars subjected to 800 o C were water-cured for 7 days after air-cooling. In order to determine the physical, mechanical and permeability properties of mortars, the bulk density, water absorption, porosity, flexural and compressive strength, capillary water absorption tests were conducted and optical microscope analyses were also performed. Results showed that mortars with MD had the highest strength values and the lowest permeability for normal and post-fire conditions, however, mortars with GCP showed the best performance among other mortars after re-curing. This can be attributed to the siliceous composition of GCP regarding the vulnerability at high temperature and suitability in C-S-H re-formation when re-curing. Water-curing contributed to re-hydration and self healing, resulting in strength re-gain and porosity recovery which was also observed by microstructural evaluation.

In this paper, Incompressible SPH (ISPH) is used to simulate free-surface mudflow for two case studies, including dambreak and flow under a gate. The mixture of water and sediment in mudflow is treated as a non-Newtonian fluid. Mass and momentum conservation equations in a 2-D Lagrangian frame, along with the Herschel-Bulkley rheology model, were solved to simulate mudflow using ISPH. Further, a Large Eddy Simulation (LES) model was used to evaluate the effect of turbulence on the free surface flow for these cases. The divergence-free velocity projection method was applied to enforce incompressibility of SPH. The results of the ISPH modeling compared well with experimental data and FVM-VOF results. The capability of the ISPH Lagrangian numerical model to capture large deformations makes it a powerful and efficient method for simulating mudflow case studies.

The time-dependent behavior of soils has been investigated extensively using one-dimensional and triaxial tests. The phenomena associated with time effects in soils are creep, relaxation, strain rate and re-arrangement effects. The engineering properties of soil are often improved significantly with the elapse of time. The objective of this paper is to investigate the time-dependent effect on the shear strength parameters of sand–geosynthetic interface using large direct shear test apparatus. For this purpose, the geotextile layer was carefully adhered to a piece of rigid block with a thickness such that half of the shear test box is occupied. The other half of the box has been filled with sand and the test performed. Three normal stresses of 30, 45, and 60 kPa have been applied in all tests. The shear stress has subsequently been applied at different times to the failure stage. In all tests, the shearing velocity has been kept the same. The results of these experiments show that the stiffness and friction angle of the sand–geotextile interface increases up to 35% and 5.5% at 720 minutes after the sample is poured in the mold. These increases occur mostly in the first two hours following the normal stress application to the samples. The findings in this paper are interesting to consider in practice.

Near-fault seismic records strongly influenced by forward directivity are characterized by one or more large pulses in velocity time histories. The characteristics of these records are significantly different from ordinary records. This study evaluated the correlation between the intensity indices of pulse-like ground motion and crest settlement of embankment dams and their ranking according to quality was evaluated. The seismic behavior of five embankment dams with different heights was investigated under 105 pulse-like and 20 ordinary ground motions with over 680 nonlinear time history analyses. The results showed that inelastic structural ground motion intensity indices are more applicable than elastic indices to predict the settlement of embankment dams. The proposed inelastic structural ground motion intensity (I MPD) calculates the mean permanent displacement of a single degree of freedom (SDOF) system over a period of 0.2T 1 to 1.5T 1 (where T 1 is the natural period of the dam) based on Newmark’s method using a decoupled approximation and deformable SDOF instead of a rigid block system. The non-structure ground motion intensity index IVA (where I VA is the square root of the product of peak ground velocity and plain integral of the squared acceleration) is proposed as a good predictor for the nonlinear response of embankment dams.

Among the various adsorbents used, sawdust appears to be the most attractive material in terms of cost, versatility and abundance. In the present research, synthetic contaminated water with gasoline was used. Adsorption of petroleum compounds on four types of sawdust (Walnut, Poplar, Beech and Pine) was studied by using batch adsorption techniques. It was found that walnut sawdust has higher adsorption capacity than other types of sawdust. It was also observed that equilibrium adsorption capacity was a function of pH, temperature, and H 2O2 concentration in solution. Maximum adsorption of petroleum compounds was obtained at pH 8. The adsorption of petroleum compounds was increased by decreasing the temperature and H 2O2 concentration in the solution. The maximum equilibrium capacity of walnut sawdust was 84.03 sawdust gr COD mg nd 606.37 sawdust gr TPH mg obtained at pH 8 and+ a temperature of 10˚C. The experimental adsorption data were fitted to a Freundlich and Langmuier adsorption model. Calculated correlation coefficients indicated that the Freundlich model was best suited, indicating that the nature of walnut sawdust is heterogeneous. In this study, to illustrate the relation between q e, C e, pH and temperature, regression analysis was taken into consideration. The obtained model (with R 2 =0.985) improved the correlation coefficient at least 1.13% compared to the Freundlich model. The maximum and minimum error of the predicted values of q e to the experimental data was obtained as 14.29% and 0.034% for regression model, while these errors for the Freundlich model were 52.15% and 1.63%.

Project selection is a fundamental decision in construction companies. Regarding the high complexity and dynamic nature of construction projects, the level of uncertainty in this industry is very high. Portfolio selection strategy can be considered as one of the best ways for mitigating the risk of project selection. With respect to the considerable importance of project portfolio selection in reducing the risk to a company, developing research in this area is crucial. The main focus of this paper is to present a hybrid model according to the fuzzy case-based reasoning for prescreening of projects according to the factors generating risk in the construction industry and the historical records of the company, also allocating the most appropriate prescreened projects to the portfolio of company with the use of zero-one linear goal programming. A real case study has been presented for model implementation and for more understanding.

Pakistan is a seismically prone country and its provincial capital city Quetta is bordered by a number of faults. Traditionally Peak Ground Acceleration (PGA) is calculated, and is further used in design calculations for structures. However, PGA alone doesnot amply define the seismic load and modern building code emphasis on use of spectral acceleration values. Seismic hazard analysis has been carried out for Quetta city and design parameters required by codes to define seismic loading have been derived. The maps are developed for 0.2 second (Ss) and 1.0 second spectral acceleration (S 1) values for 2% probability of exceedance in 50 years, equivalent to 2500 years of return period. The proposed maps have been recommended to be included in Building Code of Pakistan.

Area-capacity curves are of the most important physical characteristics of dam reservoirs. These curves are used for reservoir flood routing, reservoir operation, prediction of sediment distribution in reservoirs, etc. In this study, the linear characteristic of a reservoir’s depth-capacity data on log-log paper is used and the mathematical equation of the reservoir’s capacity-depth curve is obtained based on the shape factor, M. The reservoir’s area-depth equation and equation of shape factor are obtained by differentiating the reservoir’s capacity-depth equation. The obtained equations are evaluated with the area-capacity data of 8 reservoirs in the United States. The results of this study showed that the obtained equations agree well with the actual data.