Genetic Algorithm based Optimization of Pulse Profile for MOPA based High Power Fiber Lasers

摘要

Although the Master Oscillator Power-Amplifier (MOPA) based fiber laser has received much attention for laser marking process due to its large tunabilty of pulse duration (from 10ns to 1ms), repetition rate (100Hz to 500kHz), high peak power and extraordinary heat dissipating capability, the output pulse deformation due to the saturation effect of fiber amplifier is detrimental for many applications. We proposed and demonstrated that, by utilizing Genetic algorithm (GA) based optimization technique, the input pulse profile from the master oscillator (current-driven laser diode) could be conveniently optimized to achieve targeted output pulse shape according to real parameters' constraints. In this work, an Yb-doped high power fiber amplifier is considered and a 200ns square shaped pulse profile is the optimization target. Since the input pulse with longer leading edge and shorter trailing edge can compensate the saturation effect, linear, quadratic and cubic polynomial functions are used to describe the input pulse with limited number of unknowns(＜5). Coefficients of the polynomial functions are the optimization objects. With reasonable cost and hardware limitations, the cubic input pulse with 4 coefficients is found to be the best as the output amplified pulse can achieve excellent flatness within the square shape. Considering the bandwidth constraint of practical electronics, we examined high-frequency component cut-off effect of input pulses and found that the optimized cubic input pulses with 300MHz bandwidth is still quite acceptable to satisfy the requirement for the amplified output pulse and it is feasible to establish such a pulse generator in real applications.