Local shape optimisation of a firtree root using NURBS


Song, Wenbin, Keane, Andy J., Rees, Janet, Bhaskar, Atul and Bagnall, Steven (2002) Local shape optimisation of a firtree root using NURBS In Proceedings of the 9th AIAA/ISSMO Symposium on Multidisciplinary Analysis and Optimization. American Institute of Aeronautics and Astronautics..

Download

[img] PDF song_02.pdf - Accepted Manuscript
Download (2MB)

Description/Abstract

This paper describes a general shape optimisation method for a structural optimisation problem using a combined feature-based and B-spline based free form geometry modelling approach. The geometry is defined by a number of features for the global shape and by Non-Uniform Rational B-splines (NURBS), via the use of control points for the local shape. The coordinates of the control points along with other dimensions are chosen as design variables. A knot insertion/removal process is introduced to increase flexibility and to obtain a compact definition of the final optimised shape. Here the geometry is modelled using an intelligent CAD system (ICAD), which enables complete automation of the design-toanalysis process. Stresses are obtained from finite element analysis. A two-stage hybrid strategy is used to solve the formulated optimisation problem. Its application to the design of a turbine disc firtree root is demonstrated. This new approach is compared to a more traditional optimisation analysis using a geometry model based on straight-lines and circulararcs. Improvements in the objective function and flexibility in geometry definition are achieved compared to designs obtained using the traditional approach.

Item Type: Conference or Workshop Item (Paper)
Additional Information: AIAA 2002-5486
Venue - Dates: 9th AIAA/ISSMO Symposium on Multidisciplinary Analysis and Optimization, 2002-09-04 - 2002-09-06
Subjects:
ePrint ID: 22798
Date :
Date Event
2002Published
Date Deposited: 07 Mar 2007
Last Modified: 16 Apr 2017 22:49
Further Information:Google Scholar
URI: http://eprints.soton.ac.uk/id/eprint/22798

Actions (login required)

View Item View Item