New NWTC/NREL Fatigue-life estimation tool

Call for Alpha Testers (NWTC/NREL Fatigue-life estimation Tool)

We are happy to announce an alpha-version release of MLife, a MatLab-based tool that can efficiently process statistical and fatigue-related output needed for wind turbine certification. MLife’s calculations follow the method required by the IEC in their 61400-1 Edition 3, Annex G, design standard, and the recommendations of Germanischer Lloyd (GL) in their Guideline for Certification of Wind Turbines, 2010 edition.

The input data for MLife are typically obtained from a large series of aero-elastic simulations (FAST, for example). The software functionality of version 1.0 – including calculations of general statistics, lifetime damage, and damage-equivalent loads – has been implemented and the project is now focused on verification and validation of the tool.

Engineers at the National Wind Technology Center (NWTC) at the National Renewable Energy Laboratory had created a program called MCrunch that included these features and many more, but MCrunch must store the entire dataset concurrently in memory. For such a large number of simulations, the memory requirements would exceed the amount available in an engineer’s typical computer. Staff at the NWTC recently addressed this problem by creating MExtremes for the extreme-event tables and MLife for fatigue-life predictions. These specialized tools process files sequentially, so only one data file is stored in memory at a time.

Feel free to download the software and associated documentation from the MLife webpage:

Feedback during alpha testing will be used to identify any problems which prevent analysis and results generation in accordance to the IEC and GL recommendations mentioned previously.

You can post your comments here or email them to Greg Hayman,

Kindest Regards,

Greg Hayman
NREL subcontractor

We have discovered a bug in the v1.00.00b-gjh release related to the calculation of wind speed bin probabilities. Currently the scale factor for the Weibull distribution is set to 2*Vave / sqrt(pi), but this should only be the case when a shape factor of 2 is used (in other words a Rayleigh distribution). Therefore, you can only use a Rayleigh distribution in the current version. We will address this in the next release.

Additionally, we are considering having the user enter the distribution’s mean windspeed and standard deviation, which will in turn be used to compute the shape and scale factors of the Weibull distribution. Currently, the user enters the mean windspeed and the scale factor of the distribution.

Best Regards,

Greg Hayman


I am trying to conduct a reliability assessment on the offshore wind turbine.
I’ve read the manual for MLIFE however I am still unsure of the algorithm behind this code.
First, does this code follow Palmgren-Miner’s Linear Damage Accumulation theory?
Second, the input “DesignLife” correspond to the operation time of the Wind Turbine under investigation? (ex. 20-30 years for the typical turbine)

Thank you as always,
this seems like a very useful tool.

Kan Ito

Dear Kan,

Yes, MLife makes use of Palmgren-Miner linear damage accumulation theory.

Yes, DesignLife corresponds to the desired operational life of the wind turbine, e.g. set DesignLife to approximately 630000000 s for 20 years of operational life.

Much more information is provided in the MLife User’s Guide and Theory Manual.

Best regards,