As FLAIR image is similar to T2w image, does it can be used as structural image in DIFFPREP?
It really depends on your purpose. If your goal is to alight the DWIs onto a structural image, then yes you can safely use a FLAIR image.
However, if your goal is to also perform EPI distortion correction, that might be more difficult as the CSF signal is dark in FLAIR images compared to the b=0 image. And the metric used for EPI distortiion correction in DIFFPREP is MEanSquares (or local cross correlation) with Jacobian signal modulation, neither of which are good at significant signal differences like this.
Hope this helps.