We propose a novel hybrid genetic algorithm (GA) based on such techniques as clustering, sharing, crowding, and adaptive probability. The proposed GA can effectively solve multimodal optimization, including the global and local optima in the distributed multipump Raman amplifier (DMRA). The simulation results show that the optimal signal bandwidth Delta lambda can be evidently broadened by means of increasing the number of pumps and that Delta lambda decreases with the increase of Raman gain and the improvement of the flatness property. The optimal results show that the hybrid erbium-doped fiber amplifier and DMRA can availably overcome the weakness of pure DMRA and that both higher gain and broader bandwidth can be realized in hybrid amplifiers simultaneously.
© 2003 IEEEPDF Article