Materials Data Requirements for Computer Simulation in Design and Manufacture

Abstract

The paper considers a top-down systems approach to the wealth creation process starting at the marketplace, progressing to a design specification, the concept design phase and, through detailed design and manufacture, to delivery of the component or system to the client. At all stages of the process the need is considered for data and in particular materials data. Whilst the paper refers to the special features associated with the quality of materials data required at the concept stage of design, consideration is also given to the type of data required to underpin product innovation and technological forecasting. The bulk of the paper considers the materials data requirements for supercomputer simulation of the behavior of components in service and of products where the material is created and then used in service. The latter situation is considered via the optimization of the selection of materials for a high- temperature weldment in a pressurized steam pipe; and also via the assessment of the performance of a tension panel, containing a central circular hole, fabricated from a tough ceramic fiber-ceramic matrix composite. These examples show the diversity of requirements of materials data in supercomputer simulation at the detailed design stage. In the examples considered, reference is made to visualization and video animation to transmit two dimensional screen images of the time variation of field plots of stress and internal damage variables which can be interpreted satisfactorily by a range of cross-disciplinary specialists. Conclusions are made on the importance of linking the structure and accessibility of databases with the need to provide guidance from physics-based mechanism maps for deformation, damage and fracture.KeywordsCreep StrainHeat Affected ZoneCreep CurveMaterial DataCeramic Matrix CompositeThese 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.