Open-loop test.

[QusayHawari]

Hi,

I am trying to perform an open-loop analysis on the UMaine IEA15MW turbine model using OpenFAST.

Do you think the right way to do it is to deactivate the controller in ServoDyn by setting PCMode and VSContrl to zero, and assigning a certain blade pitch as an initial value in ElastoDyn?

Do you think it's also possible to enter an array of blade pitch commands as an initial condition instead of one value?

Thanks,
Qusay.

[jjonkman]

Dear @QusayHawari,

You can disable active pitch control by setting PCMode = 0 in ServoDyn (and by ensuring that the override pitch maneuvers are not engaged during the simulation). In this case, the initial pitch set in ElastoDyn will remain constant throughout the simulation.

Setting VSContrl = 0 in ServoDyn implies that a generator model is specified instead. Presumably by "open-loop analysis" you want to specify constant torque; is that correct? To set constant torque in ServoDyn, you can set VSContrl = 1, VS_RtTq = to the constant torque you want, and VS_RtGnSp = VS_Rgn2K = VS_SlPc = some small number greater than zero, e.g., 0.001.

Best regards,

[QusayHawari]

Hello Jason,

I' am using turbulent wind speed with a mean value of 20m/s. For the blade pitch and generator torque they should be constants since VSContrl is set at 1, and PCMode is set at 0 with an initial value of 17.9 degs for the blade pitch. Indeed they are constant looking at the results.

Qusay

[jjonkman]

And how do these plots compare between the open- and closed-loop response?

[QusayHawari]

Hi,

Here they are:

Qusay.

[jjonkman]

It looks like the response is similar between the two because the control actions are not too different between the two after the start-up transient. It appears that the closed-loop controller you are using uses constant torque in Region 3, just as in open loop.

[QusayHawari]

Yes, I agree, Since the objective of the pitch controller is to stabilize the blade pitch at the operating point of 20m/s, I am essentially doing the same thing in open-loop when I apply that objective as a form of initial value for the pitch. In both cases the torque is constant.

Thank you so much for your FAST and helpful replies :)

Qusay.