Calculating total blade nodal load

[PaulKSmith]

Hello everyone,

Is there anyway to get blade total nodal force (Aerodynamic lift + drag + centrifugal force + mass distribution) in X, Y, Z directions for N1-N40 from OpenFAST? Intention is to use the same forces as input to ED for a separate Finite Element Analysis. This is to validate my FEM model with ED. So, far I have been using (B1N1Fx, B1N1Fy,...) from AD as normal and tangential loads and adding the centrifugal load and body weight for each node based on the blade property. However, would like to know if there is feature which automatically does that. I would appreciate any help!

Regards,

[jjonkman]

Dear @PaulKSmith,

I'm not sure I fully understand your question, but AeroDyn transfers the aerodynamic applied loads (lift, drag, pitching moment) to ElastoDyn. ElastoDyn will calculate the structural response including the effects of body weight (centrifugal, weight, inertia, etc.). The AeroDyn blade nodal outputs and/or node outputs are the local aerodynamic applied loads at each cross section and the ElastoDyn blade nodal outputs and/or gage node outputs are reaction loads within the structure at each cross section (integrated from the cross section to the blade tip).

Best regards,

[PaulKSmith]

Dear @jjonkman Sorry that my question was unclear. Thank you for your reply! I just want to perform a separate Finite Element Analysis of a blade whose loads are derived from a OpenFAST simulation. For this I would need the load distribution along the blade span at each time step so that I can use it in my FE model. From your explanation, maybe I can use the reaction force at each cross-section [Spn1FLxb3, Spn1FLyb3, Spn1FLzb3] after converting it to load distribution ? Since Spn1FLxb3 ... are shear force.

Regards,

[jjonkman]

Dear @PaulKSmith,

Thanks for clarifying. Yes, I agree. You could convert the reaction loads output from ElastoDyn to a load distribution by taking the difference in reaction load between each node/station. When you do this, though, you'll have to ensure that the coordinate system of each reaction load is the same, e.g., by converting the reaction loads output in local coordinates to a consistent global coordinate system.

Best regards,

[PaulKSmith]

Dear @jjonkman ,

Thank you for the confirmation. Just want to make sure the transformation matrix from local to global coordinate system for a head wind steady case would just be the following, where the theta being the rotor azimuth angle? Do I need to account for blade pitch angle?

Regards,

[jjonkman]

Dear @PaulKSmith.

What I meant was the blade reaction loads output from ElastoDyn are expressed in a local coordinate system that is aligned with local principle axes of bending and follow the deflection of the blade. So, the coordinate transformation needs to involve the slope of the flapwise and edge deflections, as as well as the structural pretwist angle. The coordinate transformations used by ElastoDyn are documented here: https://openfast.readthedocs.io/en/main/_downloads/c9ae932057ea1e64c79154b36d3ab30a/FASTCoordinateSystems.doc.

Best regards,