Open Access Open Access  Restricted Access Subscription or Fee Access

Structural Health Monitoring: A Review of Uncertainty Quantification Methods in Wind Turbine Systems

P.H. GONZAGA, N. DERVILIS, K. WORDEN, R. BARTHORPE, N. STEVANOVIC, L.O. BERNHAMMER

Abstract


Wind energy has been in the forefront of renewable energy sources and is expected to show continuous expansion in the future. Nevertheless, there is a considerable demand for the development of new technologies as the turbines get more complicated, bigger and requirements for design, operation and maintenance become stricter. Uncertainty quantification (UQ) can contribute to further technology development, it is important to analyse both data and model results to understand limitation of current modelling, data variability and take informed decisions because uncertainties are present in all domains of wind turbine design, modelling, manufacturing and maintenance. To have adequate UQ, it is important to consider the sources of the uncertainty, which can generally be either aleatory or epistemic. Composite material with complicated microstructures have become crucial to wind turbine systems and so have their behaviour characterisation, thus requiring UQ knowledge for both the material properties and damage models. Different models and data entail ( or are available to) different UQ methodologies. With the intensification of non linear and multiphysics modelling, computational cost has also become a decisive aspect of uncertainty quantification. This article discusses the sources of uncertainties in wind turbine blade models, reviewing the main techniques used to propagate uncertainties within wind turbine systems.


DOI
10.12783/shm2019/32417

Full Text:

PDF