Using Mlife for a lifetime fatigue estimation

[Julius.Scherf]

Hi everyone,

I want to use MLife to determine the fatigue lifetime for the 5MW NREL
turbine.
So far, I have run test 18 without problems. After reading up in the forum,
I’ve to download MLife to do fatigue estimations, but I can’t find the
corresponding link.
Can somebody help me?

Thanks and best wishes,
Julius

[Jason.Jonkman]

Dear Julius,

A similar question was asked and answered in my post dated Jun 17, 2020 in the following forum topic: viewtopic.php?f=4&t=805&start=30.

Best regards,

[Julius.Scherf]

Thank you for your quick reply. That helps a lot.

Best wishes,
Julius

[Julius.Scherf]

Hi Jason,

I managed to run the mlife_certification_test file provided by mlife.
Now I want to adjust the input files in such a manner that I can determine the lifetime of the 5MW NREL turbine for all three cases specified in the IEC-61400 standard.

1. To do so, do I have to modify the DLC data files, or do I have to import the data files from FAST?
2. I’m wondering how I can plot the corresponding Wöhler curve?

Best wishes,
Julius Scherf

[Jason.Jonkman]

Hi Julius,

I’m not sure what you mean by “DLC data files”. Are you referring to the FAST time-series output files?

MLife will not plot the S-N curve for you, but you can plot it yourself using the values of L_Ult, L_MF, and m using equation (2) from the MLife Theory Manual: drive.google.com/file/d/1d7uWya … sp=sharing.

Best regards,

[Julius.Scherf]

Hi Jason,

thank you for your quick reply.
The Problem concerning the DLC files solved itself.
But regarding the S-N curve, I have a question. I understand that I can calculate the number of cycles using equation (2) with L_Ult and L_MF provided by MLife, but I’m not sure about L_i^RF. Can I understand it as such that it represents the standard deviation from each load case?

Best wishes,
Julius Scherf

[Jason.Jonkman]

Dear Julius,

L^RF_i is the cycles load range about a fixed mean value. The S-N curve is a plot of L^RF_i (S) versus N_i (N).

Best regards,

[Julius.Scherf]

Dear Jason,

thank you. I appreciate your effort.

Best wishes,
Julius

[Julius.Scherf]

Dear Jason,

I set the Weibull-Shape factor to 2, so I can assume a Rayleigh wind distribution. Further, I want to determine the lifetime damage for three different mean wind speed values (7.5; 8.5; 10 m/s). The strange thing that I observe is that the ultimate load decreases from 8.5 to 10 m/s. Is there a fundamental error I don’t see. Do you have any ideas?

Best wishes,
Julius

[Julius.Scherf]

Dear Jason,

I’m sorry, I forget to tell you that the ultimate load for the lifetime damage was calculated for RootMMxy1, which was determined as described in the MLife manual.

Best wishes,
Julius

[Jason.Jonkman]

Dear Julius,

I’m sorry, but I’m not sure I understand. How are you arriving at the ultimate load for RootMMxy1?

Best regards,

[Matthew.Lee]

Hi Jason,

I wasn’t sure where exactly to post this but it is regarding L_Ult and the use of MLife.

I am wondering what values I should insert for L_Ult in the .mlif file for MLife? The turbine I am using is the 15MW offshore wind turbine. Please let me know what you think.

Thanks and best regards,
Matthew

[Jason.Jonkman]

Dear Matthew,

LUlt has been discussed many times on our forum, e.g. see:

MLife Fatigue DEL Calculation

Dear Bruno, Unfortunately, the NWTC Information Portal has crashed and is no longer available, and there is a bit of gap before the new website becomes live–see the following announcement for more information: Other News. Regardless, here is what the old FAQ said regarding the question, “What values do I use for TypeLMF and LUlt in my MCrunch and MLife input files?”: Until the new website becomes live, I’ve temporarily places the MExtremes (as well as MCrunch and MLife) download(s) here: win…

Mlife - User Defined Distribution

Dear Conor, Just to be clear, the OC3-Hywind specifications report (nrel.gov/docs/fy10osti/47535.pdf) gives the Young’s modulus (E) of the tower as 210 GPa; however the yield strength (sigma_yield = E*epsilon_yield, where epsilon_yield is the axial strain at the yield strength) is not given in that report. I misread your post and was simply using your value. You’ll have to make your own assumptions on sigma_yield for your own purposes. Here are the basic equations for LUlt for a thin-walled …

Best regards,

[Matthew.Lee]

Dear Jason,

Thanks for your quick reply. I found the links very useful.

Best regards,
Matthew

[Andriana.Vassiliou]

Dear Jason,

I am wanting to perform a mooring fatigue life analysis using my tests from FAST V8, I have used Map++ in my simulations and wanting to know how to use mlife to process these results and work out the fatigue life.

Best regards,
Andriana.

[Jason.Jonkman]

Dear Andriana,

MLife can be used to post-process the fatigue of mooring line tensions (in terms of DELs, lifetime damage, etc.) just like it can with any other turbine or support structure load.

That said, MAP++, being a quasi-static model, does well at predicting the mean and low-frequency mooring line loads, but not the dynamic / high frequency mooring line loads. So, I would expect the mooring line tension ultimate and fatigue loads to be underpredicted by MAP++. MoorDyn is better when mooring line loads are to be analyzed.

Best regards,