Basrah Journal for Engineering Sciences

This paper deals with the behavior of reinforced concrete anchor blocks for underground steel pipelines.~ under the effect of loads caused by internal pressure and temperature variation due to the transportation of hydrocarbon products. The finite element method is used to carry out the analysis using the ANSYS 5.4 program. To study the effect of oil, it is represented by springs with different values for the modulus of subgrade reaction in normal and tangential reactions. It is concluded that increasing the values of the modulus of subgrade reactions, kn and ks of the soil surrounding the reinforced concrete anchor block causes an increase in the failure loads of the block. But at high values of these modules, the rate of this increase in the failure load will decrease. The area of the passive face of the concrete anchor block is found to have the main effect on the failure load as compared to the length of that block. The failure load of the concrete anchor blocks that have square cross sections is 1.33 times larger compared to that of rectangular sections. It is also concluded that locating the steel flange at the middle of the block leads to a larger resistance of anchor blocks as compared to any other position.

The study aimed to investigate the structural behavior of indirectly loaded flanged deep reinforced concrete beams. Twenty-one flanged deep beams were tested. The behavior of beams under loading was observed. Cracking and ultimate loads were recorded.

This paper explores the potential of using artificial neural networks to predict the ultimate moment capacity of steel-concrete composite beams with metal deck slabs. Basic information on artificial neural networks and parameters suitable for the analysis of experimental results are given. A multilayer backpropagation neural network is used for training and testing the experimental data. A comparison study between the experimental values and two models (neural network and AJSC models) is also carried out. It was found that 1he neural network model provides better results. The proposed neural network is also used to explore the effect of the various parameters on the behavior of beams.

A two-dimensional mathematical model has been constructed by using the finite difference method for the representation of the groundwater flow in both steady and unsteady states at the upper aquifer of Dibdibba formation. The hydraulic characteristics of this aquifer have been redistributed based on observed data for the period (1988• l 989). A verification test is added to check the model's correctness by matching the calculated levels with the ones observed for the year 2000. A model was set to predict the groundwater levels up to the year 2010. Results of the prediction show a reduction in groundwater level about (Im) in the central parts of the study area compared to the level of this groundwater in the year 2000. On the other hand, this decrease reaches (0.5 m) in the western parts of this area.

In this paper, a parallel optical array adder for large-scale 2D Modified Sign- Digjt (MSD) data array is proposed. and implemented to limit the cany propagation to constant steps. The digit decomposition plane (DDP) representation technique is expanded to code the 2D an-ay of the MSD number system. The design is based on the logical funnel. which are newly derived according to the fundamental parallel addition algorithm for MSD number system using the feature of the DDP coding technique. The optical implementation scheme is based on classical optical elements such as spatial light modulators, beam combiners, beam splitters, mirrors, light source analysis, and light detector arrays. The proposed algorithm and its optical architecture have useful intrinsic characteristics, such as ultra-high speed, constant processing time, and parallel computation on large-scale data arrays. The simulation results ensure that the proposed arithmetic unit is worked successfully.

This paper presents an approach for the recognition of off-line handwritten numeric strings using a genetic algorithm. The proposed scheme is divided into two parts. The first part is to remove the image noise, and then the vertical projection is used to segment the numeric strings into isolated digits and every digit will be presented separately to the second part. The second part uses an improved genetic algorithm to recognize isolated handwritten digits. The result of the recognition of the numeric strings will be displayed at the exit of the global system.

Interest in neural networks as an alternative to conventional algorithmic techniques has grown rapidly in recent years. Noise removal or noise suppression is an important task in image processing. In general, the results of the noise removal have a strong influence on the quality of the following image processing techniques. In this paper, two feed-forward NN schemes have been presented for impulsive noise removal. The computation is reduced by using an artificial image in training. Results of NN schemes show high performance especially when the ratio of impulsive noise in testing is the same or greater than that of training image. The presented schemes are used for the grayscale and also for true color.

ln this paper, image deblurring, and denoising are presented. The used images were blurred either with Gaussian or motion blur and corrupted either by Gaussian noise or by salt & pepper noise. In our algorithm, a discrete wavelet transform is used to divide the image into two parts. This partition will help in increasing the manipulation speed of images that are of big sizes. Therefore, the first part represents the approximation coefficients, that a blur is reduced b,y using the modified fixed-phase iterative algorithm. While the second part represents the detail coefficients, that a noise is removed by using the Bayes Shrink wavelet thresholding method.

The dual synchronization of two different pairs of chaotic oscillators: one pair of Duffing oscillators and one pair of Murali-Lakshmanan-Chua (ML-Chua) circuits has been achieved by numerical simulations. The cross-coupling method, where the difference in the voltage between the sum of the two master oscillators' voltages and one of the slave oscillator voltages is injected 10 the other slave oscillator as an electrical current, for the dual synchronization has been used. The accuracy of synchronization of chaos is numerically obtained by calculating the root mean square error (RMSE). A communication scheme is presented, utilizing the chaotic masking (CMS) technique. Encoding and decoding of a message based on dual synchronization of chaos has been demonstrated.

This paper describes the problem of minimizing -the side lobe levels in the radiation pattern of antenna arrays by using the genetic algorithm. Two types of genetic algorithm representation are used here: binary and continuous genetic algorithms depending on the nature of the problem at hand. Adaptive genetic algorithm which is a special type of genetic algorithm is used in this work. The obtained results explain the capability of this approach to obtain the desired side-lobe level.

This paper deals with the computer simulation of stress distribution in a plane model of mild steel under biaxial tensile loading. The goal is to visualize the crack behavior under deferent ratios of biaxial loading through linear elastic fracture mechanics theory. A finite element method is considered in calculating the mixed mode of stress intensity factor that governs the influence of stress distribution around the crack. Aspects of crack propagation are considered. It is found that the maximum circumference stress is not of the plane of the crack but that inclined by an angle (68) from it.

Al20rA l10l bronzing systems were produced using a one-stage actively brazed technique based on Cu-Ti, Cu-Zr, and Ag-Cu-Ti alloys. Single and double butt joints were used for micro-structural and mechanical properties studies respectively. The joints that were brazed by using Cu-Zr filler-metal alloys (2%, 4%, 6%, and 8% Zr weight percent) have shown low shear strengths at the 2rO2 interface. Higher shear strength was obtained by using Cu-Ti filler-metal alloys (2%, 4%, 6"/o, and 8% Ti weight percent), and eutectic (Ag-26%C u-4% Ti). As judged by the phases Conned at the interface, Cu2 (AlTi)4O is more effective to wet and both alumina to alumina.