I am currently engaged in calculating the phonon vibration frequencies of self-interstitial atom (SIA) clusters in tungsten, aiming to subsequently compute the vibrational free energy.Due to the difficulty of calculating the vibration frequencies of large defect clusters with VASP, which is not only time-consuming but also ineffective, I have to resort to using LAMMPS for the calculations.My primary concern revolves around the methodology for utilizing LAMMPS to calculate phonon vibration frequencies in systems containing interstitials, specifically SIA clusters. From my current understanding, LAMMPS is often used in conjunction with PHONOPY for phonon spectrum calculations. However, a notable challenge arises when dealing with defective structures, as the computations typically focus on perfect crystal cells and involve supercell expansion. This presents difficulties when attempting to address defects.