A CAD-based method for multi-objectives optimization of mechanical products. Issue 5 (29th July 2017)

• Record Type: Journal Article
• Authors: Castorani, Vincenzo; Vita, Alessio; Mandolini, Marco; Germani, Michele
• Is Part Of: Computer-aided design and applications. Volume 14:Issue 5(2017)
• Page Start: 563
• Page End: 571
• Publication Date: 2017-07-29
• Subjects: Design for cost -- CAD parameterization -- multi-objective optimization; Computer-aided design -- Congresses; Computer-aided design -- Periodicals; Engineering design -- Data processing -- Congresses; Engineering design -- Periodicals; 620.00420285
• DOI: 10.1080/16864360.2016.1274528
• Languages: English
• ISSNs: 1686-4360

Abstract

ABSTRACT: This paper presents an approach for the product multi-objective optimization by associating the Response Surface Methodology (RSM) and the Design of Experiments (DoE) technique with the CAD/CAE/DfC integration. The goal of this study is to empirically identify the relationships existing between the design features and the product response (stress, strain, cost, etc.). This approach integrates three different levels of analysis: optimization problem definition, virtual prototyping and design optimization. The optimization problem definition concerns in choosing the design variables, setting constrains that have to be observed and goals that have to be achieved. The virtual analysis allows defining the principal parameterization of a geometric model and simulating the performance of each configuration, at a specific working condition, while evaluating the manufacturing cost and time. The proposed approach investigates the effect of the defined parameters and noise factor on the experimental results. In particular, the applied method is based on virtual experiments according to the necessity to reduce time and costs in the early design phase. The loop of design-analysis-redesign during the optimization process is done automatically, without any interaction with the designer, by using a dedicated software tool. A test case is presented to show the characteristics of the methodology and to demonstrate its feasibility. Results demonstrate that the proposed approach significantly expedites the optimization process and reduces the computing cost compared to traditional approaches. GRAPHICAL ABSTRACT: