Abstract
This paper is a preliminary study to explore the benefits of user interaction in topology optimisation by attempting to support two distinct claims: the first claim states that there exist similarly optimal, yet visually different designs based on the exact same parameters. The second one supports that accurately predicting the outcome can guide the program to faster convergence by skipping intermediary steps. For the purpose of this research a program based on Sigmund’s 99-line MATLAB code for topology optimisation was developed to implement real-time interaction. The programming language chosen was Java® for its flexibility and ease of scripting as well as its global efficiency. Both claims were tested through two distinct sets of experiments. The first one modified the designs by adding and/or removing material and proved the existence of similarly optimal yet different designs. The second one explored the use of pseudo-filters to simulate intuition and managed significant decreases in the amount of iterations necessary for convergence. This second experiment also produced slightly stiffer designs. Both experiments led to the conclusion that user interaction, when used responsibly, helps topology optimisation in generating creativity and in speeding the process.
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
In order to be able to make a choice based on aesthetical reasons.
- 2.
See the parts on future improvements and limitations for more information on resource constraints.
- 3.
Triangles represent hinges; bars represent boundary conditions for a whole side and the thin arrows under those bars show the direction in which displacements are blocked; thick arrows represent a downwards unitary force.
- 4.
It might useful to note that the intervention of a user mid-way through the optimisation process is a special case of changing the initial start point for optimisation.
- 5.
These pseudo-filters have a role similar to actual filters but work with threshold values rather than neighbouring cells.
- 6.
Defined as “areas where the density jumps from 0 to 1 between neighbouring elements” (Pedersen et al. 2006: 1), the checkerboard patterns are a common numerical instability that occurs particularly when using Q4 elements [12].
- 7.
This case is the 4-corners case for which the solution provided by the program is far from the optimal one. It is only a marginal result and should thus be disregarded.
- 8.
This result corresponds to the one mentioned in the previous footnote.
- 9.
The reason the original design is unsymmetrical is the slight upward shift in the concentrated load. Indeed, it is not applied on the middle node but on the one above it.
References
Reiss R (1976) Optimal compliance criterion for axisymmetric solid plates. Int J Solids Struct 12:319–329
Sigmund O (2001) A 99 line topology optimization code written in MATLAB. Struct Multi Optim 21:120–127
Bendsoe MP, Kikuchi N (1988) Generating optimal topologies in structural design using a homogenization method. Comput Methods Appl Mech Eng 71:197–224
Warzee G (2010) Méthode des éléments finis (structures), 7th edn. Presses Universitaires de Bruxelles, Brussels
Nikishkov G (2010) Programming finite elements in Java. Springer, London
Bendsoe MP, Sigmund O (2003) Topology optimization. Theory, methods and applications. Springer, Berlin
Bendsoe MP (1989) Optimal shape design as a material distribution problem. Struct Optim 1:193–202
Rozvany GIN (2001) Aims, scope, methods, history and unified terminology of computer-aided topology optimization in structural mechanics. Struct Multi Optim 21:90–108
Liszka T, Ostoja-Starzewski M (2003) Effects of microscale material randomness on the attainment of optimal structural shapes. Struct Multi Optim 25:1–10
Chandrasekhar SJ, Haberb RB (1996) Stability of finite element models for distributed-parameter optimization and topology design. Comput Methods Appl Mech Eng 130:203–226
Bruns TE (2005) A re-evaluation of the SIMP method with filtering and an alternative formulation for solid-void topology optimization. Struct Multi Optim 30:428–436
Sigmund O (2007) Morphology-based black and white filters for topology optimization. Struct Multi Optim 33:401–424
Andreassen E, Clausen A, Sschevenels M, Lazarov BS, Sigmund O (2011) Efficient topology optimization in MATLAB using 88 lines of code. Struct Multi Optim 43:1–16
Tovar A, Wilson IQ, Patel NM, Renaud JE (2005) Hybrid cellular automata with local control rules: a new approach to topology optimization inspired by bone functional adaptation. In: 6th World Congress of structural and multidisciplinary optimization, 30 May–03 June 2005, Rio de Janeiro, Brazil, pp 1–11
Tovar A, Patel NM, Niebur GL, Sen M, Renaud JE (2006) Topology optimization using a hybrid cellular automaton method with local control rules. J Mech Des 128:1205–1216
dE Timary S (2011), Exploring interaction as an optimisation tool, research dissertation, adaptive architecture & computation. University College London
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer Science+Business Media Dordrecht
About this paper
Cite this paper
de Timary, S., Hanna, S. (2014). Interaction in Optimisation Problems: A Case Study in Truss Design. In: Gero, J. (eds) Design Computing and Cognition '12. Springer, Dordrecht. https://doi.org/10.1007/978-94-017-9112-0_11
Download citation
DOI: https://doi.org/10.1007/978-94-017-9112-0_11
Published:
Publisher Name: Springer, Dordrecht
Print ISBN: 978-94-017-9111-3
Online ISBN: 978-94-017-9112-0
eBook Packages: EngineeringEngineering (R0)