Dear Masih,
A similar question has been asked and answered on our forum here: User input kinematics for floater in FAST - #8 by Antonio.Pegalajar. My guess is that you are applying a first-order low- and/or high-frequency cut-offs.
Please note the WaveMod = 5 option assumes that the user-specified wave elevation is first order. Adding second-order terms along with WaveMod = 5 will not allow you to get the exact wave-elevation time series back, because additional energy will be added at the difference and sum frequencies.
Best regards,
.