Skip to main navigation Skip to main content
The University of Southampton
Economic, Social and Political Sciences

COSSAN-X: A general purpose software for the efficient uncertainty management of large Finite Element models Seminar

Social Statistics and Demography
16 May 2013
S3RI Computer Lab (Building 39)

Event details

Statistics Research Thursday Seminar Series

During recent years, usage of computer aided engineering has risen in all phases of engineering not only in design but also on e.g. life prediction, manufacturing, planning [1,2]. This has been allowed by continuous evolution of the modelling capabilities of e.g. FE software. However as accurate as these analyses are, they are not capable to completely capture the behaviour of real- life structures due to the inevitable uncertainties always present [2] in e.g. loading, materials properties and manufacturing quality. To ensure a faultless life of the products/systems and to provide decision support, the consequences of unexpected events and processes have to be considered explicitly during the design process (see e.g. [3,4]). This can only be done resorting to stochastic models for analysis and design of systems. However in this context, the utilization of such tools in practical applications remains quite limited [2]. The aim of this workshop is to demonstrate that uncertainty quantification and management can be performed on large and complex numerical models within affordable costs if the deterministic solvers are integrated efficiently within appropriate software [1]. This aim can be achieved with the general purpose software COSSAN-X that includes the state-of-the-art in stochastic analysis and simulation, provides an efficient approach to integrate third-party (deterministic) software (e.g. FE solvers), and allows high performance computing in order to reduce the analysis time. The applicability of the proposed approaches is demonstrated by means of a number of tutorials as well as case studies, of industrial interest involving detailed FE models e.g. [5-7].

[1] E. Patelli et al., General purpose software for efficient uncertainty management of large finite element models, Finite Elements in Analysis and Design 51, 2012, 31?48
[2] G. Schuller (Ed.), Computational Stochastic Mechanics, Computers & Structures - Special Issue 85 (5- 6) (2007) 233{330.
[3] P. Beaurepaire & G.I. Schuller, Modeling of the variability of fatigue crack growth using cohesive zone elements Engineering Fracture Mechanics 78, 2399-2413, 2011 [4] M. Broggi, A. Calvi and G.I. Schuller, Reliability assessment of axially compressed composite cylindrical shells with random imperfections, International Journal of Structural Stability and Dynamics, 11(2), April 2011
[5] R.E. Melchers, Structural Reliability: Analysis and Prediction, Wiley, 2002.
[6] B. Goller, M. Broggi, A. Calvi, G.I. Schuller, Efficient Model Updating of the GOCE Satellite Based on Experimental Modal Data, Computational Methods in Stochastic Dynamics, 215-235, Springer, 2011
[7] E. Patelli and M. De Angelis An Open Computational Framework for Reliability Based Optimization Proceedings of the Eleventh International Conference on Computational Structures Technology, B.H.V. Topping (Ed.), Civil-Comp Ltd. Stirling, 2012

Speaker information

Edoardo Patelli , University of Liverpool - Institute for Risk and Uncertainty. Lecturer in Uncertainty and Engineering

Privacy Settings