Integrating Theoretical, Numerical and Experimental Techniques to Master the Structural Dynamics of Aerospace Structures

Abstract eng:
As technologies become more advanced, it is inevitable that they become more sophisticated and that expectations and demands placed on them grow ever more ambitious. So it is in structural dynamics, where advances in computation techniques and facilities mean that very large and detailed mathematical models can be built and used to predict the most complex of dynamic phenomena in the most complex of engineering structures. However, sheer size (of model) is no guarantee of its fidelity, or usefulness, and so to take full advantage of these extraordinary numerical capabilities requires some subtle contributions from other domains. In fact, from experimental activities, suitably chosen so as to complement the predictive simulation processes. This lecture explores the interdependences of three key skills used in modern Structural Dynamics: Theoretical Modelling; Numerical Analysis; and Experimental Measurement – and how, by proper integration, they can provide state-of-the-art tools for the structural dynamicists, and especially for those working with the most advanced and critical structures routinely encountered in the aerospace industries. These three basic skills are nowadays combined to provide the applied techniques of Simulation and Validation and Identification, and for these, the experimental contributions are almost equally important as are the computation ones. The underlying assumption is that theoretical models will form the fundamental basis of all future design, manufacture, utilisation, maintenance and disposal and so the construction of adequate models is a primary driver. However, valid models are those which adequately describe the behaviour of a structure or a system and validity comes from the correctness of the model, rather than its size. This correctness can only come from a sound understanding of the physical behaviour that is of interest and that often requires careful (experimental) observation of the behaviour and the process of identification to capture the essential characteristics in mathematical terms. Once that has been done, the computation is more or less a matter of procedure. Later in the overall process, it is usually necessary to conduct tests to check the validity of the predictions, to ‘test’ the model upon which the predictions are based. Here again, carefully- selected experimental data can provide an invaluable confirmation, or validation, of the prediction model and thus the designs or interpretations that are based on the model. The different types of model are discussed, and a distinction is made between models which describe 'what the structure looks like' and ‘how it behaves’: the former relating closely to how we design it and the latter to whether it performs or behaves as we wish. Throughout, the emphasis is placed on the need and opportunity to integrate the different processes of theory, computation and measurement in order to be able to carry out highly cost-effective and reliable structural dynamics procedures. The talk is illustrated with examples drawn from several aerospace structures.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket