Concurrent Design and Evaluation Based on Structural Optimization using Structural and Function-oriented Elements at the Conceptual Design Phase

Abstract

Computer-aided engineering (CAE) has been successfully used in mechanical industries such as automotive industries. CAE enables us to quantitatively evaluate the mechanical performances of products and to propose an effective way to improve their performances using optimization techniques without building physical prototypes. However, CAE tools are usually utilized not at the conceptual design phase, but at the evaluation phase following the detailed design phase. This is because current CAE tools require detailed design data that does not yet exist at the conceptual design phase, and such tools also inhibit the provision of useful design suggestions that, ideally, match the way of thinking and insight of design engineers. Thus, at present, no CAE tools exist that can assist the conceptual design decision making process of design engineers. On the other hand, conceptual design processes are of great significance when seeking to create innovative and high-performance products and to shorten their development time. In order to fulfill the designer’s needs during the conceptual design phase, a new type of CAE method must be constructed, one that enables concurrent design support and evaluation, and fits the way design engineers think and explore design insights. This article presents a new structural optimization method that supports concurrent decision making so that design engineers can work to obtain innovative designs and evaluate the mechanical design details of mechanical structures at the conceptual design phase. This method is developed based on the concept of product-oriented analysis and discrete, function-oriented elements, such as beam and panel elements, since these can provide design suggestions concerning the structural evaluation of reasons as to why certain design ideas obtained are reasonable or optimal in the design sense. The basic ideas and specifications needed to construct the method are explained and the construction of the structural optimization design method is discussed. The optimization algorithm is developed using the ground structure approach and CONLIN sequential convex programming. The examples provided demonstrate the utility of the proposed methodology for supporting design engineers’ concurrent decision making, so that innovative mechanical designs can be evaluated at the conceptual design phase.

Design for construction: utilizing production experiences in development

“…sketches, blueprints, product specifications, production experiences and production cost drivers (cf. Takezawa et al, 2005). If existing processes are to be utilized, the product should be designed for the production process; however, if new processes are to be implemented, then the product and the process should be developed concurrently.…”

Collaborative multidisciplinary design optimization: A framework applied on aircraft conceptual system design

“…Multidisciplinary design optimization (MDO) is therefore considered to be an essential tool to search the large design space, including relevant disciplines simultaneously, and find the global optimum (Bi, 2013). Several MDO approaches have been proposed to address the computational challenges that arise when initiating an MDO process in different industrial applications, for example in the automotive, industrial robot, and aerospace industries (Tarkian, 2012), (Takezawa, 2005), (Amadori, 2012), (Giesing et al, 1998), (Sandberg et al, 2011) and (la Rocca et al, 2011). They consist of, for example, multiple-discipline feasible (MDF) (Kodiyalam et al, 2001), (Balling et al, 1996), All-in-ONE and individual discipline feasible (IDF), (Cramer et al, 1994), Collaborative optimization (Kroo et al 1994), bio-level integrated synthesis (BLISS) (SobieszczanskiSobieski et al 1998), concurrent subspace optimization (CSSO), (Sobieszczanski-Sobieski, 1988) and analytical target cascading (Kim et al, 2003).…”

Bi-level optimization for the cross-sectional shape of a thin-walled car body frame with static stiffness and dynamic frequency stiffness constraints

“…Therefore, it is too complicated to derive the sensitivities of these responses with respect to the design variables. However, equations (20) to (22) are all rational expressions whose computational cost is very small. Therefore, evolutionary algorithms 32 such as GAs, 33,34 simulated annealing 35 and particle swarm optimization 36,37 are very appropriate for this nonlinear optimization problem.…”

“…8,17 Saitou and co-workers [18][19][20] proposed decomposition-based assembly synthesis, where the topology and decomposition of a vehicle body-in-white is simultaneously optimized over a ground structure of the beam frame for the stiffness, the weight, the component manufacturability and assemblability. Takezawa et al 21,22 discussed a new structural optimization method, based on topology optimization techniques, using beam elements where the cross-sectional properties were treated as the design variables.…”