High-Stiffness, Lock-and-Key Heat-Reversible Locator-Snap Systems for the Design for Disassembly.

Heat-Reversible Locator-Snap System;Design for Disassembly;Parallel Genetic Algorithm;Mechanical Engineering;Engineering;Mechanical Engineering and Scientific Computing

Shalaby, Mohammed MounirSaigal, Romesh ;

University of Michigan

关键词: Heat-Reversible Locator-Snap System;    Design for Disassembly;    Parallel Genetic Algorithm;    Mechanical Engineering;    Engineering;    Mechanical Engineering and Scientific Computing;   

Others  :  https://deepblue.lib.umich.edu/bitstream/handle/2027.42/58479/mshalaby_1.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

The use of joints that can disengage with minimum labor, part damage, and material contamination is critical to ensure effective service, part reuse, and material recycling. This dissertation develops a general computational method for designing lock-and-key heat-reversible locator-snap systems that satisfy the aforementioned requirements. The lock-and-key concept is like a security code that allows easy disassembly when the right procedure is followed. It is realized by double-latching snaps that require force within a certain range to disengage, and multiple snaps that require heating multiple locations at different temperatures to disengage. During disassembly, thermal expansion constrained by locators and temperature gradient along the wall thickness are exploited to realize the deformation required to release the snaps.A generic optimization problem is posed to find the orientations, numbers, and locations of locators and snaps, and the numbers, locations, and sizes of heating areas, which realize the release of snaps with minimum heating and maximum stiffness, while satisfying motion and structural requirements. Screw Theory is utilized to pre-calculate the set of feasible orientations of locators and snaps that are examined during optimization. Multi-Objective Genetic Algorithm (MOGA) is used for solving the posed generic optimization problem. A parallel version, using manager-worker scheme, with active load balancing is developed to solve the generic optimization problem efficiently. The proposed algorithm selects between two parallelization schemes based on the average objective function evaluation time and either divides the population evenly over all processors or sends small patches of the population to the idle workers.The proposed heat-reversible locator-snap systems are applied to different case studies ranging from automotive bodies to consumer electronics. The first case study deals with joining internal frames and external panels in automotive bodies. Next, the proposed locator-snap systems are applied to a T-shaped DVD player enclosure, an enclosure model with complex mating line geometry, and a flat panel TV enclosure. In the later, the developed Parallel genetic algorithm is used and its performance is analyzed. In all case studies, the resulting Pareto-optimal solutions result in alternative designs with different trade-offs between the design objectives while satisfying all the constraints.