RemoveDrift¶

class hoomd.update.RemoveDrift(reference_positions, trigger=1)¶

Remove the average drift from a restrained system.

Parameters:

reference_positions ((N_particles, 3) numpy.ndarray of float) – the reference positions \([\mathrm{length}]\).
trigger (hoomd.trigger.trigger_like) – Select the timesteps to remove drift.

RemoveDrift computes the mean drift \(\vec{D}\) from the given reference_positions (\(\vec{r}_{ref, i}\)):

\[\vec{D} = \frac{1}{\mathrm{N_{particles}}} \sum_{i=0}^\mathrm{N_{particles-1}} \mathrm{minimum\_image}(\vec{r}_i - \vec{r}_{ref,i})\]

RemoveDrift then shifts all particles in the system by \(-\vec{D}\):

\[\vec{r}_i \leftarrow \mathrm{minimum\_image}(\vec{r}_i - \vec{D})\]

Tip

Use RemoveDrift with hoomd.hpmc.external.Harmonic to improve the accuracy of Frenkel-Ladd calculations.

Example:

remove_drift = hoomd.update.RemoveDrift(
    reference_positions=[(0, 0, 0), (1, 0, 0)]
)
simulation.operations.updaters.append(remove_drift)

Members inherited from AutotunedObject:

property is_tuning_complete¶: Check if kernel parameter tuning is complete. Read more...

Members inherited from Integrator:

Members defined in RemoveDrift:

reference_positions¶

the reference positions \([\mathrm{length}]\).

Example:

remove_drift.reference_positions = [
    (0, 0, 0),
    (1, 0, 0),
]