نشریه مهندسی عمران و محیط زیست دانشگاه تبریز
سال چهل و دوم شماره 1 (پیاپی 66، بهار 1391)

Concrete is one of the most common industrial and constructional materials due to its economical aspects. In recent years, fracture parameters of cement-based materials, such as concrete have been studied using the development of mathematical models derived from various experimental data and methods. The role of these parameters in design of structures is an important issue. In this paper, a fracture model based on artificial neural network, as an alternative way and soft computing to reduce time and calculation content, has been applied to estimate the fracture parameters of concrete (specific fracture energy i.e. the area under the complete stress- strain curve) by loading under a three-point bending (3PB) specimen. Fracture parameters of concrete, model’s output parameter, is predicted by input parameters: test age, aggregate type, maximum aggregate size (dmax), compressive-tensile strengths ratio (f´c/f´t), water-cement ratio (w/c), young’s modulus (E). Because of the fewness of data, thirty nine data were used as training data and eight data were used as validation or testing data. By using the neural network and its proper training to create logical relationship between input and output variables, an optimal model for each series of data could be made, which can be applied to evaluate the network reliability as an effective tool to estimate the fracture energy of concrete. The results showed that the mathematical models can be used to estimate the fracture parameters of cement-based materials and concrete. Also it was shown that the feed-forward multi-layer perception back-propagation artificial neural network with Lovenberg-Marquardt training algorithm and with 2 layers which has 6 and 1 neurons in the first and second layer respectively with R=0.90 and RMSE=0.04 is most appropriate neural network for the estimation of fracture energy of concrete.

Chaos theory studies complex and nonlinear dynamic systems and phenomena that their behavior seems to be stochastic and irregular at first glance, but in fact these systems are ruled by certain laws and with a deeper look, a kind of periodicity and regularity are evident in them. Sensitivity to initial conditions starting behavior, instability, non-linearity and non-periodic behavior, define a chaotic system. In addition, in chaotic hydrological systems, the short term time series can be extracted through analysis of long term time series. Furthermore, Information and relationships of the system can be achieved without any need to find the rules and governing dynamic relationships. Since the lakes water level fluctuations are dynamic and management of these ecosystems requires exact data at different intervals, chaos theory can play a unique role in acquiring information from this phenomenon. In recent years, Urmia Lake water level as an international wetland has been significantly reduced. Considering to the importance of this lake in Iran and the world, the purpose of this paper is to study the daily fluctuations in the Urmia Lake water level during the past 44 years using the concepts of chaos theory. To study the chaotic behavior of this phenomenon, correlation function, which is one of the exact methods to study these behaviors, was used; therefore, after computing the delay time and reconstructing the phase space, embedding dimension is determined using correlation function and false nearest neighbor algorithm, eventually correlation graph's slope is calculated. The non-integer digit result for this slope is an important index to identify this system as a chaotic system. In addition, studying the other indexes of chaotic behavior of water level such as positive Lyapunov exponent and broad band Fourier power spectrum show the quite chaotic behavior of time series studied.

Recently, various models have been widely used for analysis and design of the railway tracks. Since the rail is fixed to sleepers by fastening system and then track panel is installed directly on ballast, sub ballast and sub grade layers orderly without any bonding, so the superstructure is just embedded in the ballast and there are no adhesive or stuck layers. Many modeling for the purpose of analysis and design, railway track can be simulated by using linear or non-linear springs to represent the behavior of the rail on substructure. In the engineering society, the Beam on Elastic Foundation is the most well-known technique. Among these modelings and approaches like as Adin.et.al, Scott, Filonenko- Borodich foundation, Pasternak foundation, Reissner foundation, Vlasov and Leontiev approach, because of the simplicity of Winkler Model, this model hase been considered as one of the idealized models. Railway engineers and designers are most interested in choosing this modeling. Since in reality, the soil foundation does not sustain tensile stresses, the stiffness of tensioned springs should be set equal to zero in the Winkler model. In this work, regarding the advantages of simplicity and popularity of Winkler model in various engineering applications, the model has been developed by eliminating the limitations using Tsai and West Mann theory. In the modified Winkler model, the weight of rails and sleepers (self-weight) has been involved in the formulations and the springs in tensions have been disregarded during the calculations. Consequently, new set of beam-deflection differential equations have been written for each tension and compression segment and the closed form solution have been applied. Also sensitivity analysis is performed by using normalized variables. Specifically, it is shown that (i) the contact length is a sensitive function of the beam/rail profile type; (ii) also the non-contact length is depend on the magnitude of the load and that this function may change significantly regarding wheel-set load, (iii) sub grade modulus has a noticeable influence on the contact region consequently lifting-up zone length. Regarding the railway track conditions and traffic load in Iran, a case study presented, in which the length of no contact zone and the length of contact zone in the Winkler, Tsai and West Mann models have been compared. The result of rail deflections in the modified Winkler model shows that the length of no contact zone (2.5 times) and the height of track uplift due to the concentrated load (1.5 times) are greater than those obtained from the Winkler model including self-weight.

Embankments may become saturated by seepage flow during a long term high reservoir stage. If subsequently the reservoir pool is drawn down faster than the pore water can escape, excess pore water pressures and reduced stability will result. This is called drawdown which is quite common in upstream slopes of embankment dams. The stability of slopes under drawdown conditions are usually analyzed considering two limiting conditions, namely slow and rapid drawdown. In the slow drawdown situation the water level within the slope is assumed to equalize the reservoir level at any time. In case of rapid drawdown, which represents the most critical condition, it is assumed that the pore water pressure within the embankment continues to reflect the original water level. The lag of the phreatic line depends on factors such as: permeability of soils, drawdown rate, drawdown ratio and slope gradient. Rapid drawdown can cause a temporary increase in pore water pressure. The increased seepage forces may lead to slope instability, causing the collapse of structure. This paper discusses the effect of horizontal drains on upstream slope of earth fill dams during rapid drawdown using finite elements and limit equilibrium methods. The change of pore water pressure, outpouring seepage flow and the factor of safety are inspected.

The application of structural arch form in Iran goes back to as far as thousands of years ago. Arch is a fundamental component of Iranian architecture. Other structural systems such as vaults and domes which are used widely in variety of structures are indeed derived from arches. In the present study, the stability of various Iranian arches versus their weight, surcharge and earthquake action is investigated. The selected arches are some of the most applied Iranian arches like: parabolic-shaped arches, onion-shaped arches, four-centered pointed arches, obtuse angel arches and basket-handle arches. The selected arches are modeled in AutoCAD software and then are exported into the ANSYS software for necessary analyses. The static analysis, modal analysis and linear and non-linear dynamic analysis have been carried out on the finite element models of the arches. Comparison of the results of the analyses showed that for the same span and thickness in various Iranian arches analyzed in the present study, the parabolic-shaped arch has a better bearing with respect to gravity and lateral loads. The static stability of the arches under the gravity load increases with rise to span ratio of the arch. The other factor that affects the arch stability is its angle of contact to the baseline. According to the two above mentioned factors the stability of the selected arches increases in the following order: the semi-elliptical arch, the parabolic-shaped arch, basket-handle arch, circular arch and four-centered pointed arch. The eigne-value analyses have shown that the natural period of the arch in sensitive to the form of the crown. For the same span, the arches with tipped crown possess larger periods. In this regard, the fundamental period for parabolic arch is the least. This shows that this type of arch is stiffer that the other ones. For dynamic analysis of the selected arches, three strong ground motions of Kobe of Japan in 1995 (ML=7.2), Tabas of Iran in 1978 (ML=7.8) and Manjil of Iran in 1990 (Mw=7.4) were selected. Again, the time history analysis of the arches shows that the strongest arch against the selected accelerograms is parabolic arch; and the order of relative seismic stability of the arches descending from high to low is as following: the parabolic-shaped arch, the basket handle arch, the four-centered pointed arch, the Semi-elliptical arch and the circular arch. Also, under the Kobe earthquake accelerogram, the parabolic arch showed no collapse. However, the other arches encountered collapse and among the circular arch collapsed at 5.8 seconds as the first arch and the basket handle arch collapsed at 8.5 as the last one. As another result, the location of cracks in the arches damaged by seismic loading was in the internal face of the arch.

Concrete structures in aggressive environments greatly suffer from corrosion which causes a premature failure in their life span. Persian Gulf region is one of the most aggressive environments in which concrete structures need further attention in order to increase their durability and serviceability as well. In the present investigation the half-cell potential and corrosion resistance of rebar embedded in concrete with different water to cement ratios exposed to splash zones in Qeshm Island for 18 month was studied. The results demonstrated that concretes with lower water to cement ratio perform better in terms of corrosion protection and process retardation.