Sextractor in the same aperture, aperture photometric flux and Sex + PSFex PSF flux difference of more than 30%

[xiaoguaishoubaobao]

When I measure the psf img_r.fits flux with sextractor+psfex In Phosim Decam Data,
aperture photometric flux and Sex + PSFex PSF flux difference of more than 30% of all object ,
img_r.list is three object center position.
set_0 and set_1 is flux difference of more than 30%

The link is the data and configuration file :
share url_1:https://app.box.com/s/e9o25ry3nysc8p45siakp3agwmfsdnxn
share url_2：https://pan.baidu.com/s/1z-dB4xgZ7tDjtE_C6cTIng pwd：hplu

Is this a bug?
If it's not a bug ,
How should I modify sextractor+psfex to get the accurate psf fluxes?
Or how to modify the generated simulation data to get the accurate psf flux?

Thank you very much!