Search results

This paper aims to describe the development of an advisory system that helps building sound finite element (FE) models from computer-aided design data, with actual uncertainty levels expressed by error values in per cent, as today there is no widely accepted tool for FE idealisation error control.

The goal is to provide a computer-aided engineering (CAE) environment which assists the FE modelling phase. A demonstration program has been developed that leads the user through a step-by-step process and helps to detect idealisation errors. Uncertainties are identified and analysed following the procedure. An example illustrates the methodology on the collapse analysis of aerospace stiffened panels.

The design shows how a knowledge-based system can be used to aid a safe virtual product development.

The extension of current CAE environments is difficult, as the programs do not provide sufficient flexibility, changeability and FE solver independence. New developments can take the presented concept as a starting point.

The application of error control strategies increases the FE modelling fidelity and can prevent incorrect design decisions. The practical conversion of FE idealisation support depends on the ambitions of CAE software providers.

This research shows how a previously paper-and-pencil-based error control procedure can be transformed to an easy-to-use tool in modern software.