Question about velocity of system center of mass in fix deform + nvt/SLLOD command

Hello!

I combine the Kremer-Grest model with fix deform + nvt/SLLOD command

Just like

fix 1 all nvt/sllod temp 1.0 1.0 1.0 tchain 1
fix 2 all deform 1 xy erate ${Rate} units box remap v

and other codes are same as the input script in bench folder

In manual, it is said the stream velocity has been removed.
However, I found the velocity of system center of mass in flow direction, i.e. the sum all particle’s, is not “zero”, it increases with strain and is related to shear rate

So, I want to confirm if I miss some details in my script or it is the truth in this problem

By the way,
if combine the Kremer-Grest model with DPD, and fix comands are changed by
fix 1 all nve
fix 2 all deform 1 xy erate ${Rate} units box remap v

I also found the velocity of system center of mass in flow direction is not “zero”, but it is not increases with strain

So, if someone talk about this problem before?

Sincerely
Yongjin Ruan