[gmx-users] Conserved energy ("Conserved En.") in NVT simulation

[Wade]

Dear Mark and ALL,
   I am trying to calculate the conserved energy (H-tilde) in Bussi's stochastic velocity rescale algorithm (jcp,2007).
In previous maillist, ‍Mark had mentioned that the conserved En. in ener file is the H in bussi's paper (http://permalink.gmane.org/gmane.science.biology.gromacs.user/63005‍‍). 
According to the eq. 15 in Bussi's paper, we just need to get the time integration of dK if we have had the H.
I carefully checked the coupling.c file, and noticed a variable - therm_integral[i] in the fucntion vrescale_tcoupl().
The therm_integral[i]-= Ek_new - Ek. It seems like an accumulation of the -dK.
If it realy is the sum of -dK, the problem could become simple.
What we need to do is just like this: H-tilde=H-Tau-t* therm_integral[i].?
Is that right? I really need your help.
Any suggestion is valuable.

Wade‍