Journal of Aerospace Science and Technology
Volume:4 Issue: 1, Spring 2007

The buckling analysis of rectangular laminated composite plates with an edge delamination under in-plane compressive loading is performed using the finite element method. Such a plate may be considered as a simplified model of stiffener plates of a stiffened panel. The buckling load and buckling mode are obtained by solving an eigenproblem. In an unconstrained analysis, physically inadmissible modes appear because of the overlap between separated sublaminates in the delaminated region. To eliminate overlap, constraints are added iteratively on the entire overlapped area using penalty function method. The validity and superiority of the analysis method in predicting buckling load and buckling mode is shown in comparison with the experimental and analytical results available in the literature. Numerical results show the effect of delamination width and depth and boundary conditions on the buckling load.

The fracture and crack growth of mechanical structures is a usual phenomenon which is due to the application of tensile, cyclic loading or thermal stresses on the structure. So introducing of methods to prevent the crack growth is useful. Afterward, one of the repairing methods of crack growth, consisting to make a hole in the crack tip to elimi-nate the sharp corners, was explained. This method is frequently used in air and space industry. The drilling can be done in various locations. Three locations in the crack tip for the hole were considered. In the first, the hole was situated in the left of the crack tip (Location A). In the second one, it sited in the center of the crack tip (Location B) and in the last one; it sited in the right site of the crack tip (Location C). The study of resistance against crack growth was based on the comparison of these three systems. We have analyzed the influence of theses stopping holes, their diameters and their lo-cations on the fracture toughness. The specimens were made of Al 7075-T6 alloy and they had the shape of the Compact Tension specimen. The solicitation was a mono-tonic tensile loading in the Mode I of fracture. The stress intensity factor and the criti-cal load versus to the crack length of the test bars were computed. Numerical method was carried out by finite element method using Ansys software. Following this nu-merical analysis, fracture experimentation was carried out on the tensile machine to evaluate the influence of the stopping hole on the critical load of initiation of the crack growth. In result the best location of the hole increased 54% the critical load and the location which gived the weakest result showed 11% of increasing of the load. Thus in result of this research the best location of the stopping hole and its diameter were ob-tained to increase the life of the mechanical pieces.

In this study, the results of parallelization of a 3-D dual code (Thin Layer, Parabolized Navier-Stokes solver) for solving supersonic turbulent flow around body and wing-body combinations are presented. As a serial code, TLNS solver is very time consuming and takes a large part of memory due to the iterative and lengthy computations. Also for complicated geometries, an exceeding number of grid points are required that results in larger serial computation times. Therefore parallelizing this code would bring about a large saving in computer time and memory. In this study, a cluster of 16 computational nodes with 2.4 and 2.8 GHz, P4 CPU has been used. Also MPI library is used for communicating data among processors. Domain is partitioned in a 1-D form in longitudinal, radial and circumferential directions and results are compared with those of serial computations. There are several methods for data communication among processors such as blocking send and non-blocking send. The performance of each method is evaluated and the best method for the problem at hand is determined. The results are compared in terms of run time, speed-up and efficiency for executing the parallel code on 1, 2, 3, 4, 8, 12 and 16 processors. Also the parallel results are compared with serial results and the correctness of the parallel code is proved for each case. The effect of different partitioning direction and their interaction with the turbulence modeling is studied and the best choice is shown. The limitations of using Baldwin-Lomax turbulence model in a parallel program are discussed and a remedy is presented.

This paper studies the application of an inverse methodology for problem solving in fracture mechanics using the finite element analysis. The method was applied to both detection of subsurface cracks and the study of propagating cracks. The procedure for detection of subsurface cracks uses a first order optimization analysis coupled with a penalty function to solve for the unknown geometric parameters associated with the internal flaw. The objective function is calculated from normalizing the finite element determined displacements by the prescribed ones at some arbitrary points of the damaged component. The technique was also used for determination of the propagating both 1-D and 2-D planar crack growth directions using the well known maximum strain energy release rate criterion. In all cases studied, a good agreement between the theoretical and/or the experimentally observed crack behavior and the developed technique is achieved.

This paper presents the effect of preload on static and dynamic performance characteristics of several gas-lubricated noncircular journal bearing configurations. The linearized system approach using finite element method is used to obtain both steady state and dynamic characteristics. The results of the investigation shows that preload has a significant effect on frictional power loss and the stability margin. It is observed that increased preload, lowers the stability margin while it will increase the amount of power loss in noncircular gas lubricated bearing systems. The results also show that the preload effect on stability margin is more pronounced at lower compressibility numbers.

An unsteady two-dimensional finite-volume solver was developed based on Van Leer’s flux splitting algorithm in conjunction with “Monotonic Upstream Scheme for Conservation Laws (MUSCL)” limiters and the two-layer Baldwin-Lomax turbulence model was also implemented. To validate the solver, two test cases were prepared and the computed results had good agreements with reference data. The rotating-stall-like (RS) effect in a multi-blade 2-D stage of an axial compressor was investigated. The RS was captured with a 40% reduction in flow coefficient and a 0.4% increase in load coefficient with respect to normal operating condition. The velocity traces showed a periodic behavior during RS. The same behavior was observed with a stator-free approach, but with different modal characteristics. Finally, the RS vortices and its flow characteristics were observed in detail, and the stator-free approach seemed to be more adequate in stability margin determinations. The same observation is finally prepared for variable number of blades to show the dependency of the RS modal characteristics to number of blades, and to find the minimum required number of blades in numerical analysis of RS.