Is there any study about how the time step in wind file affects fatigue load? I think different time step will cause different fatigue load though the average wind speed, turbulence intense and random seed are same.
I am not aware of any papers directly on this topic. If you are getting different results, you should use a smaller step.
You should also note that even if the random seed is the same, the number of random variates will not be the same when you vary the time step. You will get different phase content in the generated time series. Thus you really can’t compare the fatigue loads without doing many simulations using each time step (keeping all other values constant).
I suppose one could generate a TurbSim output file at a high frequency and write a program to read the file and generate a new one with fewer time steps to test this.
Marshall