A theory is presented to examine the formation of wrinkles in plates when subjected to high rates of loading in the axial direction. The type of instability examined occurs in metals when the strains are well beyond the elastic range. For this reason the metals are assumed to be governed by the equations of a rigid plastic material. In particular, the von-Mises yield criterion is used in conjunction with the Levy-Mises flow rule. A parameter is introduced which measures the lateral restraint of the plate. By giving this parameter different values, all plate widths can be examined. The theory predicts wavelengths of the buckled plates which are compared with some experimental results obtained in 1968 by Goodier. / Applied Science, Faculty of / Mechanical Engineering, Department of / Graduate

A bar cell in the shape of an isosceles trapezoid is devised for use in flexure problems of circular plates. The cell members are endowed with elastic properties such that it deforms in the same manner as a piece of plate in condition of arbitrary uniform bending about any axis in the plane of a plate. The stiffness matrix of the cell is derived explicitly. Two methods of computation of stresses are described, one by the nodal forces and the other by the nodal displacements. The validity of the cell is tested on two examples whose exact solutions are known. One involves a semicircular clamped plate under uniform load, and the other a simply-supported circular plate under an eccentric concentrated load. The results compared favourably with the theory of elasticity solutions and the no-bar finite element solutions. Good trend of convergence of solutions is indicated on reduction of the element size. / Applied Science, Faculty of / Civil Engineering, Department of / Graduate

The incremental structural analysis program NISA83 was used to investigate various parameters affecting the ultimate capacity of square plates with circular perforations subjected to uniform shear stress. Both nonlinear material properties and nonlinear geometry were taken into account in determining the ultimate in-plane capacities and buckling capacities of perforated shear plates. The parameters investigated during this study were the hole size for a concentric location, and the hole location for a constant ratio of hole diameter to plate width of 0.2. In addition various doubler plates were studied to determine the most effective shape to restore a shear plate to its original ultimate in-plane capacity. For the first two parameters, the analysis was separated into three parts. The ultimate in-plane capacity, elastic buckling capacity and the ultimate elastic-plastic buckling capacity was determined for each combination of the two parameters. These were used to identify the importance of both elastic buckling and nonlinear material contribute to the reduced ultimate plate capacities. The results from plates with a concentrically located hole of varying size showed excellent correlation with other published experimental and analytical results for both the in-plane capacity and the 3-dimensional buckling capacities. Variation of the center location of a hole of a standard size provided some significant results. Little change was found in the ultimate in-plane capacity for all hole locations. On the other hand, the elastic buckling capacity was raised by 50% after moving the hole from the plate tension diagonal to the compression diagonal. Finally, from the ultimate elastic-plastic' buckling capacity results it was concluded that the concentric provides lower bound capacity for all other hole locations. The in-plane analysis of the optimum doubler plate size showed wide and thin plates to be more effective than narrow and thick plates. A doubler plate with the same thickness as the plate and twice the diameter of the hole is recommended to restore the perforated plate to its original in-plane capacity. In order to aid in the tedious task of checking the input data and to provide a convenient way of displaying the result, a full graphic post-processor was developed as part of this thesis. The program NISPLOT used color graphics available at the UBC Civil Engineering lab to process the output from NISA83. It was written in FORTRAN 77, utilizing subroutines from a commercial graphics package, DI3000, to obtain device independent graphics. NISPLOT generated plots of the nodes and element mesh for each data check. When a complete analysis was carried out by NISA83, nodes, element mesh, deflected shape, and color stress fill plots were generated. / Applied Science, Faculty of / Civil Engineering, Department of / Graduate

This thesis presents an experimental study and numerical predictions of the separated-reattaching flow around a bluff rectangular section. This laboratory configuration, chosen for its geometric simplicity, exhibits all main features of two-dimensional flow separation with reattachment. Detailed turbulent flow measurements of the mean and fluctuating flow field are reported. The measurement techniques used are: hot-wire anemometry, pulsed-wire anemometry and pulsed-wire surface shear stress probes. The separated shear layer appears to behave like a conventional mixing layer over the first half of the separation bubble, but exhibits a lower growth rate and higher turbulent intensities in the second half. In the reattachment region, the flow is found to be highly turbulent and unsteady. A finite difference method is used, in conjunction with a modified version of the TEACH code, to predict the mean flow field. Two discretization schemes are used: the hybrid-upwind differencing (HD) scheme, and the bounded-skew-hybrid differencing (BSHD) scheme. Laminar flow computations are performed for Reynolds numbers in the range 100 to 325. The HD computations underpredict the separation-bubble length by up to 35% as a result of false diffusion. The BSHD predictions, on the other hand, are in excellent agreement with the experimental results reported in the literature. Turbulent flow computations using the ƙ - ∈ turbulence model and the BSHD scheme result in a reattachment length about 30% shorter than the present measured value. When a curvature correction is incorporated into the model, a reattachment length of 4.3.D, compared to the experimental value of 4.7D, is predicted. The predicted mean flow, turbulent kinetic energy field and pressure distribution are in good agreement with experimental observations. An alternative method of analysis, based on the momentum integral technique, is presented. The method is not applied to the blunt-rectangular plate problem, but its use is illustrated for the simpler case of the flow in a sudden expansion, and promising results are obtained. / Applied Science, Faculty of / Mechanical Engineering, Department of / Graduate

The purpose of this investigation was to develop an eighteen parameter flat triangular finite element for analyzing plate and shell structures. The development of the element was accomplished by combining a plate bending element with a new plane stress element. The well known nine parameter triangle using the normal displacement and two slopes at each vertex was used for the plate bending element. This element contains an incomplete cubic for the normal displacement. For the in-plane element, complete cubics were used initially for the displacements and then various constraints were imposed to reduce the number of generalized co-ordinates to nine, namely the two in-plane displacements and an in-plane rotation at each vertex. One of the constraints, namely that the included angle at each vertex was invariant, destroyed the completeness of the element. However, the element was compatible in the plane. A patch-type test of the in-plane element showed that it could not represent all constant strain states exactly. However, the errors were small. The complete element was then tested on a plane stress cantilever beam, a square plate subjected to membrane stresses only, a cylindrical shell, a spherical shell and a non-prismatic folded plate structure. In all cases, reasonable engineering accuracy was achieved with modest grids of elements. Thus it was concluded that the incompleteness of the in-plane element was not too important. Finally, a compatible beam element was formulated and tested to supplement the triangular element. The beam element formulation included unsymmetric crosssections. / Applied Science, Faculty of / Civil Engineering, Department of / Graduate

Traditional metallic materials lead steel-reinforced concrete structures to a durability problem due to its low value of resistance to corrosion. The superior performances of FRP, including the high resistance to corrosion, the flexible and complex shapes… give it a big advantage. However, premature failure due to debonding of adhesives between concrete and reinforcing materials is the major concern for all types of reinforcement containing FRP plate reinforcement. This thesis gradually develops three elastic foundation models, which are mainly derived from the solution of superficial stress in the foundations-soil system. The one-parameter Winkler’s elastic foundation model is simple and easy. The two-parameter elastic foundation model thinks over the interfacial shear force of the joint bond. And the three-parameter foundation model additionally considers the adhesive layer’s transverse displacement to meet the boundary condition of zero shear stress. Finite element analysis (FEA) is used to compare with the proposed three foundation methods.

No description available.

No description available.

No description available.

No description available.