Decomposition-Based Assembly Synthesis of Space Frame Structures Using Joint Library

[+] Author and Article Information
Naesung Lyu

Department of Mechanical Engineering, University of Michigan, Ann Arbor, MI 48109-2125

Kazuhiro Saitou1

Department of Mechanical Engineering, University of Michigan, Ann Arbor, MI 48109-2125kazu@umich.edu


Corresponding author.

J. Mech. Des 128(1), 57-65 (Nov 25, 2004) (9 pages) doi:10.1115/1.1909203 History: Received June 13, 2004; Revised November 25, 2004

This paper presents a method for identifying the optimal designs of components and joints in the space frame body structures of passenger vehicles considering structural characteristics, manufacturability, and assembleability. Dissimilar to our previous work based on graph decomposition, the problem is posed as a simultaneous determination of the locations and types of joints in a structure and the cross sections of the joined structural frames, selected from a predefined joint library. The joint library is a set of joint designs containing the geometry of the feasible joints at each potential joint location and the cross sections of the joined frames, associated with their structural characteristics as equivalent torsional springs obtained from the finite element analyses of the detailed joint geometry. Structural characteristics of the entire structure are evaluated by finite element analyses of a beam-spring model constructed from the selected joints and joined frames. Manufacturability and assembleability are evaluated as the manufacturing and assembly costs estimated from the geometry of the components and joints, respectively. The optimization problem is solved by a multiobjective genetic algorithm using a direct crossover. A case study on an aluminum space frame of a midsize passenger vehicle is discussed.

Copyright © 2006 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Figure 15

Definitions of the beam cross-section design and welding design: (a) (thicki,0,thicki,1)∊FSi, the ith beam cross-section design used to define variable y (upper/lower thickness thicki,0 and side thickness thicki,1 and (b) (weldi,0,weldi,1)∊Wi, the ith potential joint location welding design used to define variable z (weld thickness for component 0, weldi,0 and for component 1, weldi,1)

Grahic Jump Location
Figure 14

(a) Resulting 42 basic members and (b) constructed structural topology graph with 42 nodes and 66 edges

Grahic Jump Location
Figure 13

(a) Possible joint locations and (b) feasible joint types at each location

Grahic Jump Location
Figure 12

(a) ASF for Audi A8 (20) and (b) simplified frame model used in the case study. Extruded and cast components are shaded dark and light, respectively.

Grahic Jump Location
Figure 11

(a) Sample design, (b) three components with three bends (shown as shaded spheres), and (c) four cast sleeves for joining

Grahic Jump Location
Figure 10

Estimating torsional spring rates using FE analysis

Grahic Jump Location
Figure 9

(a) Detailed 3D solid model of a joint and (b) its FE model with plate elements for beams, solid elements for casting, and plate elements for welding

Grahic Jump Location
Figure 8

(a) Frames with connected by a T-joint and (b) their FE model with beam elements (solid lines) and torsional spring elements (gray lines)

Grahic Jump Location
Figure 7

(a) Three subgraphs and (b) corresponding three components

Grahic Jump Location
Figure 6

Selected joint types and topology graph with the corresponding edges removed

Grahic Jump Location
Figure 5

(a) Seven basic members (m0–m6) and (b) structural topology graph with seven nodes (n0–n6) and ten edges (e0–e9)

Grahic Jump Location
Figure 4

Specified six potential joint locations (0–5, colored as gray boxes) and possible configurations (types) for each joint

Grahic Jump Location
Figure 3

(a) Sample space frame structure and (b) decomposed structure with joints (welded cast sleeves)

Grahic Jump Location
Figure 2

Approaches used in this paper

Grahic Jump Location
Figure 1

(a) Audi A2 and (b) ASF (20)

Grahic Jump Location
Figure 18

Convergence trend of the optimization run for case study using 100 numbers of generation with 1000 numbers of population. Note that after a certain number of generation, number of individuals in the Pareto set converges to about 70% of the entire population.

Grahic Jump Location
Figure 17

Radial basis networks. In this case study, R=6, S1=1250 and S2=1 (adapted from (35)).

Grahic Jump Location
Figure 16

Loading and boundary conditions (33)

Grahic Jump Location
Figure 20

Designs selected in the Pareto results in Fig. 1: (a) Design A (14 components) with good stiffness and assembleability, (b) design B (12 components) with good manufacturability, and (c) design C (12 components) balanced in all three objectives

Grahic Jump Location
Figure 19

(a) Pareto solutions at generation=100. Total 656 individuals among 1000 population are in the Pareto set. (b) Pareto solutions near fassm=−57.5. (c) Pareto solutions near fmfg=−380.0. (d) Pareto solutions near fstiff=−2.4×10−3. Utopian points are located at the upper right corner of each graph.



Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In