At the nanoscale, it depends on the material structure, if you are testing a single crystal, bicrystal, polycrystal, or another material structure the results could change. Lammps is an excellent code to perform MD simulations and dislocations can be visualized with Ovito.
Direct MD allows you to calculate dislocation nucleation rate, but only at stresses close to the threshold values. You need to create such a stress-temperature state (NPT) and wait until the dislocation nucleates. The nucleation rate can be found by averaging the statistics of nucleation times, which should be a Poisson distribution.
Free energy methods such as umbrella sampling, metadynamics, or neb could be used to find activation parameters in a wider range of stresses. Many of them are available in LAMMPS additional packages (plumed, colvars). I also recommend you to look at "fix pafi" command in LAMMPS (Swinburne PRL 2018), which uses neb results to find activation parameters at finite temperature.
Thankyou Ilya Bryukhanov CAN YOU PLESE TELL ME HOW TO USE PLUMED AND COVLARS IN LAMMPS FOR THESE CALCULATION OR WHERE I CAN FIND THESE RESOURCES. I WILL BE REALLY THANKFUL TO YOU.. AND IS THERE ANY WAY I CAN CONTACT YOU.