questions about reference control input

This forum if for discussing controls. Questions about how to implement controls in FAST are more appropriate to the CAE Tools forum.

Moderator: Bonnie.Jonkman

Jason.Jonkman
Posts: 3753
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: questions about reference control input

Postby Jason.Jonkman » Fri May 04, 2018 2:19 pm

Dear Cheng,

I wouldn't say that you could design a controller to force the acceleration of platform pitch to zero, but you could design a controller to reduce the oscillation in acceleration.

Best regards,
Jason Jonkman, Ph.D.
Senior Engineer | National Wind Technology Center (NWTC)

National Renewable Energy Laboratory (NREL)
15013 Denver West Parkway | Golden, CO 80401
+1 (303) 384 – 7026 | Fax: +1 (303) 384 – 6901
nwtc.nrel.gov

Cheng.Zhang
Posts: 43
Joined: Tue Nov 14, 2017 6:50 am
Organization: ECN
Location: france

Re: questions about reference control input

Postby Cheng.Zhang » Wed May 23, 2018 7:25 am

Dear Jason,

I have some questions concerning the FAST.
I designed a pitch controller and applied it on FAST v7 with 2 activated DOFs (GenDOF and PtfmPDOF), it works well on the SIMULINK. But when I activated ALL DOFs on FAST, the SIMULINK cannot run. The errors are as the figure following.

Could you tell me where the problem is? Thank you.

Best regards,
Cheng
Attachments
123.JPG
123.JPG (117.61 KiB) Viewed 1476 times

Jason.Jonkman
Posts: 3753
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: questions about reference control input

Postby Jason.Jonkman » Wed May 23, 2018 9:54 am

Dear Cheng.Zhang,

Warnings regarding a “small angle approximation violation,” warnings regarding “supersonic blades,” a simulation crash, or very large deflections that occur in the time series near the start of a simulation are good signs of a numerical instability. Without knowing more about your simulation settings, it's hard to know that what the problem is. But these issues have been discussed many times on this forum. I would use "Search..." in the upper-right corner to see how others have resolved similar issues in the past.

Best regards,
Jason Jonkman, Ph.D.
Senior Engineer | National Wind Technology Center (NWTC)

National Renewable Energy Laboratory (NREL)
15013 Denver West Parkway | Golden, CO 80401
+1 (303) 384 – 7026 | Fax: +1 (303) 384 – 6901
nwtc.nrel.gov

Cheng.Zhang
Posts: 43
Joined: Tue Nov 14, 2017 6:50 am
Organization: ECN
Location: france

Re: questions about reference control input

Postby Cheng.Zhang » Wed May 23, 2018 10:22 am

Dear Jason,

Thank you for your response.

Could you tell me that when activating all the DOFs in FAST, is there a big impact on the controller performance?
And, can the blade pitch angle be negative? Are there saturations on the blade pitch angle?

Best regards,
Cheng

Jason.Jonkman
Posts: 3753
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: questions about reference control input

Postby Jason.Jonkman » Wed May 23, 2018 1:56 pm

Dear Cheng,

Yes, the structural DOFs can impact the performance of the controller.

The pitch angle can be negative, depending on definition of aerodynamic twist (the sum of the pitch + twist angles equals the angle between the chord and rotor plane). The twist is often defined such that "optimal" pitch is near zero degrees. It is common for the minimum blade-pitch angle to be saturated at the optimal pitch.

Best regards,
Jason Jonkman, Ph.D.
Senior Engineer | National Wind Technology Center (NWTC)

National Renewable Energy Laboratory (NREL)
15013 Denver West Parkway | Golden, CO 80401
+1 (303) 384 – 7026 | Fax: +1 (303) 384 – 6901
nwtc.nrel.gov

Cheng.Zhang
Posts: 43
Joined: Tue Nov 14, 2017 6:50 am
Organization: ECN
Location: france

Re: questions about reference control input

Postby Cheng.Zhang » Wed Jun 13, 2018 7:48 am

Dear Jason,

Recently, I met a problem about linearization.

First, I used CBP control,
and I used 2 DOFs(GenDOF, PtfmPDOF) linearized model before.
Now, I want to activate more DOFs and carry out a more precise linear model by FAST.
But, I found that when activating the DOFs on blades (e.g. FlapDOF1, EdgeDOF...), the linear model's states on blades can not coincident with these on the FAST model.
I knew if I use IBP, I should use MBC3 to transfer the rotating states on no rotating frame, but in my case, I used CBP as the input.

So my question is, how to linearize a model from FAST with FlapDOF1 and EdgeDOF DOFs enabled while using CBP as input.
Could you tell me how to deal with this problem?

Thank you!

Best regards,
Cheng

Jason.Jonkman
Posts: 3753
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: questions about reference control input

Postby Jason.Jonkman » Wed Jun 13, 2018 8:39 am

Dear Cheng,

Yes, you can enable the blade flap and edge DOFs together with a collective-blade pitch control input. I'm not sure what problem you are running into, but in this case, the multi-blade-coordinate transformation will apply to the blade DOFs (which are in the rotating frame), but not the collective-blade control input (which is in the fixed frame).

Best regards,
Jason Jonkman, Ph.D.
Senior Engineer | National Wind Technology Center (NWTC)

National Renewable Energy Laboratory (NREL)
15013 Denver West Parkway | Golden, CO 80401
+1 (303) 384 – 7026 | Fax: +1 (303) 384 – 6901
nwtc.nrel.gov

Cheng.Zhang
Posts: 43
Joined: Tue Nov 14, 2017 6:50 am
Organization: ECN
Location: france

Re: questions about reference control input

Postby Cheng.Zhang » Wed Jun 13, 2018 8:54 am

Dear Jason,

Thank you for your fast reply.
I'm not sure but do you think whether it's right like this:

Apply MBC to blade DOFs, but not the collective-blade control input. That is to say, the state space model is like this,

delta_Xdot_nr = A_nr *delta_X_nr + B*delta_u
delta_Y_nr = C_nr*delta_X _nr+ D*delta_u

the subscript nr means use mbc3 to transfer the AvgAMat to MBC_AvgA; AvgCMat to MBC_AvgC, and mbc3 do not work on matrix B and D.

Best regards,
Cheng

Jason.Jonkman
Posts: 3753
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: questions about reference control input

Postby Jason.Jonkman » Wed Jun 13, 2018 8:59 am

Dear Cheng,

The multi-blade-coordinate (MBC) transformation applies to any states, state derivatives, inputs, or outputs that are in the rotating frame. Because you have states in the rotating frame (and hence state derivatives in the rotating frame), linear matrix B will be effected by MBC (at least by premultiplication on the left-hand side). You haven't said what your outputs are, but if you have outputs in the rotating frame, linear matrix D will also be effected by MBC.

Best regards,
Jason Jonkman, Ph.D.
Senior Engineer | National Wind Technology Center (NWTC)

National Renewable Energy Laboratory (NREL)
15013 Denver West Parkway | Golden, CO 80401
+1 (303) 384 – 7026 | Fax: +1 (303) 384 – 6901
nwtc.nrel.gov

Cheng.Zhang
Posts: 43
Joined: Tue Nov 14, 2017 6:50 am
Organization: ECN
Location: france

Re: questions about reference control input

Postby Cheng.Zhang » Wed Jun 13, 2018 9:16 am

Dear Jason,

First, the outputs including rotating DOFs.

I agree that xdot and x should apply MBC, but the input u is CBP control command, it is in no rotating frame.
What I confused is how to apply MBC on state alone, and does not affect the input.

Do you think it's right to apply MBC to get A_nr=MBC_AvgA, C_nr=MBC_AvgC and leave B = AvgBMat, D = AvgDMat?

Thank you!

Best regards,
Cheng

Jason.Jonkman
Posts: 3753
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: questions about reference control input

Postby Jason.Jonkman » Wed Jun 13, 2018 9:19 am

Dear Cheng,

The MBC3 software will apply the necessary transformations directly. If you have states and outputs in the rotating frame, you should use MBC_AvgB and MBC_AvgD rather than AvgBMat and AvgDMat.

Best regards,
Jason Jonkman, Ph.D.
Senior Engineer | National Wind Technology Center (NWTC)

National Renewable Energy Laboratory (NREL)
15013 Denver West Parkway | Golden, CO 80401
+1 (303) 384 – 7026 | Fax: +1 (303) 384 – 6901
nwtc.nrel.gov

Cheng.Zhang
Posts: 43
Joined: Tue Nov 14, 2017 6:50 am
Organization: ECN
Location: france

Re: questions about reference control input

Postby Cheng.Zhang » Wed Jun 13, 2018 9:34 am

Dear Jason,

1. I used mbc3.m got matrix MBC_AvgA, MBC_AvgB, MBC_AvgC, and MBC_AvgD.
2. The input is CBP angle. (a step signal)
3. Output including BldPitch1, BldPitch2, BldPitch3...

Then, BldPitch1 shoud = BldPitch2 = BldPitch3 = input. but the result shows that input = BldPitch1. BldPitch2 and BldPitch3 are almost 0.

I don't know the reason...

Best regards,
Cheng

Jason.Jonkman
Posts: 3753
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: questions about reference control input

Postby Jason.Jonkman » Wed Jun 13, 2018 11:37 am

Dear Cheng,

It sounds like you've applied the MBC transform so that while the linear model generated by FAST has individual outputs BldPitch1, BldPitch2, and BldPitch3, once you apply the MBC transform, these outputs are transformed to BldPitch_collective, Bld_Pitch_cosine, and Bld_Pitch_sine. So, the response you are getting is expected.

Best regards,
Jason Jonkman, Ph.D.
Senior Engineer | National Wind Technology Center (NWTC)

National Renewable Energy Laboratory (NREL)
15013 Denver West Parkway | Golden, CO 80401
+1 (303) 384 – 7026 | Fax: +1 (303) 384 – 6901
nwtc.nrel.gov

Cheng.Zhang
Posts: 43
Joined: Tue Nov 14, 2017 6:50 am
Organization: ECN
Location: france

Re: questions about reference control input

Postby Cheng.Zhang » Wed Jun 13, 2018 1:29 pm

Dear Jason,

Yes, you are right, I applied MBC transform.
So the MBC just transfer the 3 blade pitch angles outputs to BldPitch_collective, Bld_Pitch_cosine, and Bld_Pitch_sine. It does not affect the input. That is to say, the linear model I got from FAST is right? Even the output blade pitch angles are different, I still can use CBP as an input?

Actually, I compare the two linear model, one is applied MBC3 and another is not. I got the almost same output from these two models, it's strange.
furthermore, I also compare these two models' outputs with the FAST model (In FAST, I used the inverse MBC transform to the rotating output), both the two linear models' outputs are coincident with FAST nonlinear model.
It surprised me. I don't know which model I should use...

Best regards,
Cheng

Jason.Jonkman
Posts: 3753
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: questions about reference control input

Postby Jason.Jonkman » Wed Jun 13, 2018 4:17 pm

Dear Cheng,

So the MBC just transfer the 3 blade pitch angles outputs to BldPitch_collective, Bld_Pitch_cosine, and Bld_Pitch_sine. It does not affect the input. That is to say, the linear model I got from FAST is right? Even the output blade pitch angles are different, I still can use CBP as an input?

Correct.

Actually, I compare the two linear model, one is applied MBC3 and another is not. I got the almost same output from these two models, it's strange.
furthermore, I also compare these two models' outputs with the FAST model (In FAST, I used the inverse MBC transform to the rotating output), both the two linear models' outputs are coincident with FAST nonlinear model.
It surprised me. I don't know which model I should use...

I'm not sure I know enough about the details to comment.

Best regards,
Jason Jonkman, Ph.D.
Senior Engineer | National Wind Technology Center (NWTC)

National Renewable Energy Laboratory (NREL)
15013 Denver West Parkway | Golden, CO 80401
+1 (303) 384 – 7026 | Fax: +1 (303) 384 – 6901
nwtc.nrel.gov


Return to “Controls”

Who is online

Users browsing this forum: No registered users and 1 guest