TNS · Developing the Next Generation Framework for Testing Nonlinear Dynamic Structures.

Developing the Next Generation Framework for Testing Nonlinear Dynamic Structures.

The constant drive to improve engineering structure performance is increasingly leading to lighter and more flexible designs where nonlinearity is inherent. Though frequently ignored or overlooked at the design stage, nonlinearity is found to affect the dynamics of a significant number of structures and its effects are commonly observed during test campaigns. Nonlinear systems can exhibit a wide range of complicated phenomena inherently very difficult to predict. However, until now, there has been no general-purpose systematic method that can directly measure and characterise such nonlinear dynamic behaviour during laboratory tests; hence it is extremely challenging to incorporate nonlinear features into the model development and validation process. Control-based continuation (CBC) is a systematic method designed to fill this void. The method uses sensors and actuators to intelligently probe a physical system using a feedback control system in conjunction with numerical continuation algorithms that track different types of behaviour as the system inputs are changed. In this way, CBC modifies, on-line, the excitation signal in order to directly interrogate the nonlinear dynamical features of interest, thus offering the best conditions to analyse them in detail. There is the potential to apply CBC to dynamic experiments across the breadth of engineering but, while demonstrating great promises, the method lacks robustness and cannot currently be applied to real-world applications. The aim of this research is to turn CBC into a more general framework for testing nonlinear dynamic structures. In particular, the algorithms currently used within CBC are ideal for mathematical models as they can be cheaply evaluated to high precision. Neither of these benefits is realised for an experiment where measurement noise is present. As a result, experiments could only be performed in ideal environments with low noise levels. In this project, a novel continuation algorithm that is robust to noise was developed and demonstrated experimentally by tracking the limit-point bifurcations of a nonlinear oscillator with adjustable nonlinearity. The project also pioneered the application of the method to wind tunnel experiments.

Data: CORDIS, © European Union