Journal of Aerospace Engineering

The results of a detailed study of the effect of cutout on the nonlinear response of curved unstiffened panels are presented. The panels are subjected to combined temperature gradient through-the-thickness combined with pressure loading and edge shortening or edge shear. The analysis is based on a first-order, shear-deformation, Sanders-Budiansky-type shell theory with the effects of large displacements, moderate rotations, transverse shear deformation, and laminated anisotropic material behavior included. A mixed formulation is used with the fundamental unknowns consisting of the generalized displacements and the stress resultants of the panel. The nonlinear displacements, strain energy, principal strains, transverse shear stresses, transverse shear strain energy density, and their hierarchical sensitivity coefficients are evaluated. The hierarchical sensitivity coefficients measure the sensitivity of the nonlinear response to variations in the panel parameters, as well as in the material properties of the individual layers. Numerical results are presented for cylindrical panels and show the effects of variations in the loading and the size of the cutout on the global and local response quantities as well as their sensitivity to changes in the various panel, layer, and micromechanical parameters.

The instability problem of imperfect, laminated, moderately thick, circular cylindrical shells under the action of individual and combined application of uniform axial compression and external lateral pressure is investigated. The analysis is based on nonlinear kinematic relations, where the effect of transverse shear deformation is taken into account. The buckling is assumed to be elastic and the geometry to have initial geometric imperfections. A solution methodology has been developed and employed in generating results. The imperfection sensitivity is investigated. The results obtained indicate that geometric imperfections have small effect on the limit point load for moderately thick cylindrical shells. In addition, limited parametric studies have been made in order to assess the effect of length-to-radius ratio, radius-to-thickness ratio, and stacking sequence on the critical loads.

This paper presents a second-order sensitivity analysis for a smart structural system formulated in terms of multiobjective optimization where structural design variables and control parameters are equally treated as design variables. The analytical derivatives of various objectives and constraints are obtained. These expressions provide a complete sensitivity package for the multidisciplinary analysis and optimization problem. An example truss regulated using four actuators is presented as verification of the derivations. This includes the optimal designs achieved using sequential linear and quadratic programming methods. Numerical results show the advantages of using high-order derivatives for the solution of a problem of such complexity and confirm that a systematic sensitivity analysis is essential, as the performance of the controller relies heavily on the structural design variables as well as feedback control logic.

Removable and reusable sensors are essential in applications like health monitoring or the nondestructive evaluation of newly produced or finished composite structures. In such tests, a major requirement is that the tested structures are not altered during either bonding or removal of the sensor. In this paper, removability and reusability have been introduced to polyvinylindine fluoride (PVDF) film sensors by a very simple process that uses double-sided adhesive tapes. The repeatability of signals generated by such sensors is verified through a series of tests. The use of these removable and reusable sensors is then demonstrated through a selected application where PVDF sensors are ideally suited, namely, detection of defects such as impact damage, saw cuts, and delaminations in elastic structures. These defects change the local curvature in beams, plates, or shell structures, and PVDF sensors are ideally suited to detect these changes in curvatures.