About NREL 5MW FOWT fatigue analysis

[axxyyyyyyyy378]

Hello all，

Now I am using MLife to calculate the short term DEL of mooringline tension of NREL 5MW FOWT. I've read the Mlife theory manual and the user guide (but don't fully understand) and done a short-term DEL calculation based on a case, but I have some questions about some of the parameters in the .mlif file.

Q1: In my case, the load condition on the FOWT is uniform wind (wind speed is 8 m/s) and irregular waves. Under this load, the FOWT is in the normal operation, and the wind speed is evenly distributed in the whole wind field (although it does not accord with the reality) , so I would like to know in this load condition what is the role of the"Distributions" section in the .mlif file, how it works to caculate the short-term DEL ? Should i use Turbsim to create my wind and make it more realistic？

Q2: in the"Fatigue" section, how can i determine the value of"EquivalentFrequency"? In"Theory manual", the product of “elapsed time” and "EquivalentFrequency" determines the denominator of the DEL calculation formula, so the value of "EquivalentFrequency" is closely related to the result.

[jjonkman]

Dear @axxyyyyyyyy378,

Here are my answers to your questions:

1. The Distributions section of the MLife input file is used when you are extrapolating the simulation results to the full lifetime of the wind turbine when computing lifetime fatigue damage. The Distributions section would not be used if you are only computing short-term damage equivalent loads, which are computed for each individual time series. In wind turbine loads analysis, it is generally better to use full-field turbulence (e.g., from TurbSim) rather than uniform wind inflow.
2. The MLife input EquivalentFrequency is the frequency of the constant amplitude cycle of the damage-equivalent load. Often this is taken to be 1 Hz.

Best regards,

[axxyyyyyyyy378]

Dear @jjonkman

Thanks for your reply. It has given me a better understanding of this issue. I apologize for asking again, but could you provide more explanation about the EquivalentFrequency in the MLife input file, which is usually considered to be 1Hz? I'm a bit confused about this. I previously suspected that its value might be related to the frequency of the external load. But now I guess that this value is just a parameter used for calculating and evaluating the fatigue damage of wind turbines. The frequency of the external load does not directly determine the value of EquivalentFrequency. Is this correct?

Best regards,

[axxyyyyyyyy378]

Dear @jjonkman ，

I am thinking another question and would like to hear your suggestions. The question is as follows: If a new mooring line is designed for a type of floating wind turbine (such as a semi-submersible floating wind turbine), how should the superiority of the new mooring line be scientifically evaluated? What quantitative parameters should be used to evaluate this superiority? What I can think of is that under the same external load, the floating wind turbine has a smaller six-degree-of-freedom displacement and the tension of the mooring line after using the new mooring line. But can we use some more specific data? For example, short term DEL?"

Best regards,

[jjonkman]

Dear @axxyyyyyyyy378,

I typically see EquivalentFrequency set equal to 1 Hz, but you could certainly check the sensitivity of the damage-equivalent loads to this value. Changing EquivalentFrequency would change the number of equivalent cycles a given simulation would have. I would guess it would make the most sense to set EquivalentFrequency equal to some representative frequency of the system, but most literature I see uses 1 Hz regardless of the wind turbine.

I would guess cost is the most important factor to determine mooring line "superiority", although that is not a direct output of OpenFAST. OpenFAST can be used to identify whether a given mooring line is sufficiently designed by computing the ultimate and fatigue loads the mooring line will experience, which can be compared to the strength/resistance of the line.

Best regards,

[axxyyyyyyyy378]

Dear @jjonkman ,

Thanks for your reply, which has given me a deeper understanding of these issues. I have a further question: In your previous answer, you mentioned that ultimate/fatigue loads can be used to evaluate the superiority of mooring lines, but you did not mention the evaluation of wind turbine displacement. I am thinking that under the same external load, if a mooring line can limit the displacement of the wind turbine within a larger range, can it be considered worse? (As shown in the figure below, I applied a turbulence wind based on turbsim and a set of regular wave loads to the wind turbine. In the surge degree of freedom, mooringline1 significantly increased the motion amplitude of the floating wind turbine. Can it be considered that its performance is worse?) Based on this, I am thinking of a method to quantitatively evaluate the superiority of the mooring line, such as proposing a conceptual formula: Mooring line “superiority” = coefficient * motion amplitude control + coefficient * mooring line fatigue load + coefficient * mooring line ultimate load. What do you think of this idea, is it scientific?

Best regards,

[jjonkman]

Dear @axxyyyyyyyy378,

I wouldn't say that designing a mooring system to limit FOWT motion within a given watch circle is of key importance, unless there are specific considerations that necessitate limited surge/sway motions such as avoiding nearby obstacles or neighboring turbines or preventing nonlinearities in mooring system response that may drive system loads to increase. (Of course, limiting rotations such as platform pitch is likely important, but pitch is only heavily influenced by moorings in taut-line systems).

But I'm not an expert on mooring system design. Perhaps others with more expertise in that area (such as @mattEhall) would like to provide further comments.

Best regards,

[axxyyyyyyyy378]

Dear @jjonkman,

Thanks for your reply, I have seriously considered your suggestions. Recently, I want to learn the source code of OpenFAST, but I seem to have not found reliable reference. Could you please give me some tips for learning the OpenFAST source code? Also, do you know any reliable resources for learning, something like "user guide"?

Best regards,

[jjonkman]

Dear @axxyyyyyyyy378,

OpenFAST user and developer documentation is available here: https://openfast.readthedocs.io/en/main/source/dev/index.html#dev-guide.

Though a little out of date, the NWTC Programmer's Handbook provides a solid foundation for understanding the architecture of the OpenFAST source code, including the registry, module structure, and programming guidelines: https://openfast.readthedocs.io/en/main/_downloads/2a9110b5e2a2b531c7dc05ea1fc86b89/NWTC_Programmers_Handbook.pdf.

Best regards,