Unexpected aerodynamic performance of IEA 15 MW turbine when using BeamDyn with DBEMT_Mod=2

[honorhan]

Dear OpenFAST developers,

I am using OpenFAST to simulate the IEA 15 MW Onshore turbine, and I encountered a possible issue when using BeamDyn together with the dynamic BEM model.

The simulation information is:

OpenFAST version: v4.1.2
Turbine model: IEA 15 MW fixed-base Onshore equivalent
Blade structural model: BeamDyn (CompElast = 2)
Controller: ROSCO v2.10.1
Wind condition:
steady uniform wind
wind speed = 8 m/s
no wind shear
no turbulence

AeroDyn settings:

Wake_Mod = 1
BEM_Mod = 1
UA_Mod = 3
DBEMT_Mod = 2

When using:

CompElast = 2 (BeamDyn)

DBEMT_Mod = 2

the turbine does not reach the expected operating point.

The final results are approximately:

Rotor speed: 5 rpm
TSR: around 8
Generator power: about 4.2 MW

However, according to the reference performance of the IEA 15 MW turbine, the expected values at 8 m/s steady wind should be approximately:

Rotor speed: about 5.7 rpm
Generator power: around 6–7 MW
I performed several comparison cases.

Case 1:
CompElast = 2 (BeamDyn)

DBEMT_Mod = 2

The operating point is incorrect.

Case 2:

Keeping all other inputs unchanged and only changing:

DBEMT_Mod = 0

The results become reasonable:

rotor speed returns to the expected value;
power output is restored;
TSR, thrust and torque are reasonable.
Case 3:

Using:

CompElast = 1 (ElastoDyn)

DBEMT_Mod = 2

The turbine performance is also reasonable.

Could you please advise:

1.Is this behavior a known issue?
2.Are there any known limitations when using BeamDyn with DBEMT_Mod=2?
3.I would like to continue using DBEMT_Mod = 2, but I have encountered an issue. Could you please advise how to solve this problem?

[jjonkman]

Dear @honorhan,

This issue sounds related (or identical) to what has been discussed in the following OpenFAST issue: #983.

I'm curious if upgrading to OpenFAST v5 and using the new tight-coupling solver resolves the issue?

Best regards,