Physical-driven fault prediction of high-power fiber lasers

Based on the characteristics of the reflected optical power and pulse signals accompanying the generation of the stimulated Brillouin scattering (SBS) effect, this paper presents a signal feature recognition algorithm. Feature extraction and identification are carried out based on the pulse data captured during experiments, and a numerical model is established. After both simulation and experimental verification, this model can accurately capture the pulse signals generated due to the SBS effect during the laser output process, thereby enabling fault prediction for fiber lasers.