Dear Shijing Xu,
I’m not an expert on Bladed, but I would expect you to get similar, but not identical, results between FAST and Bladed. Perhaps the differences you are seeing are the result of small differences in the input file settings between the models are different assumptions made by each software.
The tower-top yaw moment plot you are showing is odd to me because the sign of the mean moments are different; I would not expect this, which suggests some difference in the input file settings.
The difference in tower-top pitch moment could be the result of a different aerodynamic thrust or a different tower-top mass or center-of mass location.
The ElastoDyn module of OpenFAST does not consider mass or elastic offsets from the pitch axis, but the BeamDyn module of OpenFAST does. But it is difficult for me to guess why the pitching moment you are showing differs.
To investigate further, I would probably start by simplifying the model further. Do you get the same static deflections and loads when you disable the aerodynamics and load the wind turbine by gravity only?
Please see the following forum topic for more information on the theory basis of the ElastoDyn module of FAST: Coupled blade modes in FAST.
Best regards,