Abstract
A full-scale experimental test for large and complex structures is not always achievable. This can be due to many reasons, the most prominent one being the size limitations of the test. Real-time dynamic substructuring is a hybrid testing method where part of the system is modelled numerically and the rest of the system is kept as the physical test specimen. The numerical–physical parts are connected via actuators and sensors and the interface is controlled by advanced algorithms to ensure that the tested structure replicates the emulated system with sufficient accuracy. The main challenge in such a test is to overcome the dynamic effects of the actuator and associated controller, that inevitably introduce delay into the substructured system which, in turn, can destabilize the experiment. To date, most research concentrates on developing control strategies for stable recreation of the full system when the interface location is given a priori. Therefore, substructurability is mostly studied in terms of control. Here, we consider the interface location as a parameter and study its effect on the stability of the system in the presence of delay due to actuator dynamics and define substructurability as the system’s tolerance to delay in terms of the different interface locations. It is shown that the interface location has a major effect on the tolerable delays in an experiment and, therefore, careful selection of it is necessary.
Highlights
For many engineering structures, full-scale physical tests are not achievable; for example, if the size of the structure is large or the related costs make the test impractical, or because not all of the components are available for testing at the appropriate point in the design process
One could test individual substructures of the full system, it cannot be guaranteed that the substructure always behaves in exactly the same way when it is part of an overall structure, neither can the behaviour of the overall structure be deduced if there is nonlinear coupling between the substructures
The sensor is responsible for sending the force feedback to the numerical model, and the actuator imposes the interface displacement, which is computed by the numerical model, to the physically tested component
Summary
Full-scale physical tests are not achievable; for example, if the size of the structure is large or the related costs make the test impractical, or because not all of the components are available for testing at the appropriate point in the design process. Residual phase error or lag remains [12] It is common for this phase lag to be modelled as a constant delay; this is considered accurate because, typically, the excitation frequency band in a substructuring test is narrow and so the frequency dependency of the actuator dynamics is negligible. Examples of where this approximation has been used include [1,2], and its validity is further supported by successful applications of delay compensation techniques to reduce it This allows us to show the effect of the interface location on the stability of a substructured system more generally
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call