HNEMD Convergence test

[achyuttt]

Hello everyone,

I was doing HNEMD for some of my research work for which I wasn't able to perform GreenKubo calculations due to convergence issues.

I am concerned on how to know if HNEMD results are converged ? My same potential gave me way too different converged results when I did HNEMD in different sizes with different driving forces.

Does HNEMD have size effects?

Do we need to check the different driving force convergence?

Like HCACF in GK, is there a way to check if the HNEMD thermal conductivity should be picked up from a certain region?

Please let me know if anybody have idea on these questions.

[brucefan1983]

There many examples in the literature. The most important thing is that you need to perform a few independent simulations to get an error estimate in any case.

To discuss further, it is better to present input scripts and results.