Date of Award

2011

Level of Access

Campus-Only Thesis

Degree Name

Master of Science (MS)

Department

Mechanical Engineering

Advisor

Senthil S. Vel

Second Committee Member

Zhihe Jin

Third Committee Member

Alireza S. Sarvestani

Abstract

Sandwich panels have long been known to exhibit high specific strength and stiffness, leading to substantial design advantages over conventional monolithic plates. Properly designed sandwich panels offer substantial resistance to static and dynamic loads due to their high relative stiffness and inherent energy absorbing capacity. To that end, steel sandwich construction has great potential for use in ships, building and bridge structures, especially for hazard reduction in situations of high wind, storm surge, earthquakes or accidental blast. Early investigations of sandwich construction show great promise but have highlighted the need for more advanced methods of connecting the core to the facing sheets. Laser welding techniques hold great promise for overcoming the manufacturing impediments of the past allowing for the large scale production of sandwich panels for a number of different applications. The sandwich panel core is an essential element in sandwich panel construction, used predominantly to resist transverse shear force, like the web of an I-beam. Core designs for sandwich panels can take on a number of different forms depending on the intended end use. A sandwich panel designer must choose a set of geometric parameters, that define the sandwich panel, from a multitude of potential choices. Since the structural response and weight of a sandwich panel is highly dependent on the choice of geometric parame-ters, it is important to develop a robust methodology for the optimization of laser-welded steel sandwich panels. In order to obtain optimal sandwich panel designs for an intended end use, we present a methodology for the multi-objective optimization of laser-welded steel sandwich panels subject to static and blast loading. We utilize an integer-coded non-dominated sorting genetic algorithm with multiple, conflicting objectives to obtain an entire Pareto-optimal front of equally optimal sandwich panel designs. As opposed to a single optimal design, an entire front of optimal designs allows for trade off studies by the designer in order to find the most applicable sandwich panel. Candidate sandwich panel designs are analyzed using a geometrically nonlinear finite element analysis in the general purpose finite element package ABAQUS. The utilization of a finite element package for design optimization provides the designer with the ability to investigate local effects in geometrically complex designs where analytical methods may not. Results are presented through multiple model optimization problems for static and blast loads. An optimization is performed for static loads on a non-optimized reference geom-etry given in the literature in order to highlight the effectiveness of the presented opti-mization methodology. Selected optimal designs are compared with the non-optimized reference geometry to investigate the benefits of sandwich panel optimization. Optimiza-tion results are also presented for square laser-welded steel sandwich panels subjected to different static load magnitudes, objective functions, and fixed number of prismatic cores. The final model problem methodology and results are presented for the optimization of square laser-welded steel sandwich panels subjected to air blast loading. It is shown that the combination of a genetic algorithm and a general purpose finite element package can provide a robust and effective method of finding optimal sandwich panel designs for mul-tiple, conflicting objectives.

Share