Loading...

A Comparative Study of Numerical Integration Algorithms Used in Real-Time Hybrid Simulation

Rezazadeh, Hassan | 2013

806 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 45121 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Ahmadizadeh, Mehdi
  7. Abstract:
  8. Real time hybrid simulation is considered as one of the most efficient experimental methods for investigation of the behavior of structures and their components during seismic loading. In pseudo-dynamic approach of hybrid simulation considered in this study, the majority of the structural mass is modelled numerically and inertia forces are calculated in the numerical model.Using this approach, loading can be carried out on a real-time basis, and component scaling requirements will be reduced. As a result, in addition to lower costs, this testing method is sometimes considered to be more realistic compared to shaking table tests. However, several challenges are currently faced in performing an accurate real-time hybrid simulation, one of them being the selection of suitable integration algorithms for this purpose. This is due to the fact that both numerical and experimental models and errors are present in hybrid simulations. This has led to the development of several modified versions of integration algorithms that were originally introduced for purely numerical simulations. This work is a comparative study of the existing numerical integration methods for hybrid simulation that are developed in recent years. These investigations are carried out in the presence of numerical and simulated experimental errors.In addition, an improved numerical integration procedure is introduced by combination of two existing methods, namely the implicit Newmark integration algorithm and the integration method with combined implicit or explicit steps. It has been demonstrated that the new proposed integration algorithm results in a reduction of energy balance error in low to moderate experimental noise
    levels
  9. Keywords:
  10. Laboratory Bench Mark ; Hybird Simulation ; Integration Algorithm ; Numerical Damping ; Stability ; Accuracy

 Digital Object List

 Bookmark