Developing a Finite Element Model in Conjunction with Modal Test for Wind Turbine Blade Models

Abstract

There are numerous modeling approaches that can be employed in the generation of composite models for wind turbine applications. Several more traditional approaches have been used over the past few decades. However, one modeling approach used for the generation of manufacturing models uses what is referred to as the unit cell approach. This modeling approach has been used for large deformation applications where the orientation realignment of the composite plies needs to be considered in the forming process. This modeling approach has been used for static and large deformation applications but not for dynamic modeling scenarios. There would be a significant advantage to use this approach for the complete modeling of wind turbine blades. A comparison of the traditional composite ply modeling approach and the unit cell approach is studied to determine similarities and differences in the approaches. A panel section is modeled and tested in various conditions followed by correlation and updating studies. A comparison of the different modeling approaches is performed.The use of this modeling technique is specifically intended to be used in conjunction with the determination of dynamic stress–strain prediction for turbine blade applications. A simple break-out section of a turbine blade comprised of balsa and resin/fiber is used to identify typical model properties needed to properly model the turbine blade. A panel section is modeled and tested and then the model is updated to reflect the properties to best represent the panel structure. The test is performed with several different perturbed boundary conditions to assure that the updated model has appropriate and realistic properties identified for the model.KeywordsMode ShapeComposite PlateComposite LayerComposite PanelOrthotropic PlateThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.