Dear @Lin.Ding,
I’m glade MLife is now working as you expect.
Good question regarding combining bending moments and axial forces to predict axial stress and strain. MLife only works directly with the loads, and so, axial force is treated separately from pure bending even though both create axial stress and strain. To do fatigue properly requires post-processing in terms of stress or strain based on the cross-sectional properties of the component. This can be done analytically for simple cross sections composed of isotropic materials, as has been discussed in the following topic on our forum: Mlife - User Defined Distribution. I know such analytical treatment of stress/strain from loads capability is being developed within pCrunch (GitHub - NREL/pCrunch), which is python based (NREL has been transitioning from MATLAB to python-based pre- and post-processors.) But for complicated cross sections such as airfoils composed of composite materials, a sectional analysis tool is needed.
Best regards,