Innovative Approach to Achieve Topology Optimized Designs More Efficiently

This presentation was made at the NAFEMS European Conference on Simulation-Based Optimisation held on the 15th of October in London.

Optimisation has become a key ingredient in many engineering disciplines and has experienced rapid growth in recent years due to innovations in optimisation algorithms and techniques, coupled with developments in computer hardware and software capabilities. The growing popularity of optimisation in engineering applications is driven by ever-increasing competition pressure, where optimised products and processes can offer improved performance and cost-effectiveness which would not be possible using traditional design approaches. However, there are still many hurdles to be overcome before optimisation is used routinely for engineering applications.

The NAFEMS European Conference on Simulation-Based Optimisation brings together practitioners and academics from all relevant disciplines to share their knowledge and experience, and discuss problems and challenges, in order to facilitate further improvements in optimisation techniques.

Resource Abstract

Classical topology optimisation is a well-known and proven method since a long time helping to find improved part designs. With the emerging possibilities of additive manufacturing in the last years it became more and more possible also to manufacture these designs without compromises at an affordable effort.



Nevertheless, classical topology optimisation techniques have some serious constraints. One is that the standard approach is to look for the stiffest design at a given weight target. This has two disadvantages: first of all, the weight is not the result of an optimised design but needs to be prescribed as input. This means that the lightest possible design is not found automatically. Secondly the found design is not stress proved and its strength needs to be checked separately. If the stress is too high, manual geometry adaptations and further check loops are necessary.



Another severe limitation is that the classical topology optimisation does not deliver a final part design, but just an indeterminate idea in terms of a blurred density field. For sure this is a great help on the way to an optimised design, nevertheless interpretation and (semi-)manual derivation of a distinct design is necessary, increasing the workflow effort significantly. Even smoothing algorithms have difficulties in transferring this density field into a tangible geometry, as there is no unique solution. Also, a direct and reliable check of the stress state is not possible unless an explicit design is defined.



In order to overcome these issues of classical topology optimisation, an innovative approach has been developed and practically proved. Still based on finite element simulation it waives using a density field rather than utilizing a very fine mesh with well-defined elements. This enables the algorithm to directly evaluate the occurring stresses, as well as reliably derive a distinct geometry. The advantages are obvious: it is possible to directly find an optimal structure that fulfils the stress requirements at the least possible weight. Furthermore, a clearly defined geometry is found that can be directly used and even automatically transferred into generic CAD formats. All in all, this leads to a dramatic decrease of effort in the design optimisation workflow. It becomes obsolete to manually derive a design and perform additional stress check and design adaptation loops.



Especially this helps to make the utilization of the rather expensive additive manufacturing technology worthwhile, as large design efforts at the beginning of a workflow can be saved and thus driving down the total costs and lead time per part. Amongst other examples the development of a formula student wheel carrier as a practical use case is presented that proves the new efficient way of working.

Document Details

ReferenceC_Oct_19_Opt_26
AuthorMehmert. P
LanguageEnglish
TypePresentation
Date 15th October 2019
OrganisationSimufact Engineering GmbH
RegionUK

Download


Back to Search Results