University of Twente Student Theses


Uncertainty quantification in flexible multibody dynamics

Voorthuizen, K.L. van (2020) Uncertainty quantification in flexible multibody dynamics.

[img] PDF
Abstract:Simulations of flexible dynamic systems are generally performed under the assumption that the input variables, such as the geometric and material properties, are exactly known. In actuality, a number of geometric and material properties are uncertain. To determine the influence of the uncertain parameters on the output of the model, one can sample the distributions of each uncertain parameter a large number of times, evaluate the deterministic dynamic model for each sample and use the obtained samples of the output parameter to estimate its distribution. However, simulations of flexible dynamic systems typically are expensive. Evaluating the simulation for each sample is therefore not feasible. By reduction of the model, the evaluation time might be reduced to such extend that the large number of evaluations can be performed within an appropriate time frame. The deterministic model can be reduced through physics-based methods or by a data-driven approach to reduce the number of unknown variables in the model. Alternatively, the stochastic model can be reduced by discretization of the stochastic space to obtain a metamodel which approximates the original system but consists of polynomials which are inexpensive to evaluate. This report shows a successful application of different data-driven reduction techniques on deterministic and stochastic flexible dynamic systems.
Item Type:Essay (Master)
Faculty:ET: Engineering Technology
Subject:52 mechanical engineering
Programme:Mechanical Engineering MSc (60439)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page