摘要
提出并验证了基于随机并行梯度下降(SPGD)算法的主动相位控制脉冲激光相干合成方案。利用低通滤波器滤除脉冲激光光强变化导致的性能评价函数起伏,提取出相位噪声导致的性能评价函数变化,然后将这个低通滤波后的性能评价函数用于SPGD算法的极值寻优过程,从而实现脉冲激光的锁相。对该方案进行了理论分析和数值模拟,并进行了两路脉冲激光相干合成实验。实验结果表明,当系统从闭环到开环时,系统性能评价函数均值提高到开环的1.6倍,干涉条纹长曝光对比度从0提高到0.43。在该方案中,通过增加合成脉冲激光路数,并在各路脉冲激光中引入多级功率放大器,能够得到更高的合成功率输出。
A scalable architecture for coherent combining of pulsed lasers with active phase control using stochastic parallel gradient descent (SPGD) algorithm is proposed and demonstrated. A low-pass filter is introduced to eliminate the fluctuation of the metric function caused by pulsed lasers and extract the exact phase noise signal, and an active control is thereby built with SPGD algorithm and stable coherent beam combining can be obtained at the receiving plane. Theoretical analysis and numerical simulation were made and an experiment of coherent beam combining two-channel pulsed fiber lasers was also conducted. Experiment results shows that the power encircled in the main-lobe is 1.6 times increased and the fringe visibility of the long exposure pattern is increased from 0 to 0. 43 when the system evolves from open-loop to closed-loop. This architecture can be easily scaled up to high power by increasing the number of channel for coherent combining and adding power amplifiers in each channel.
出处
《国防科技大学学报》
EI
CAS
CSCD
北大核心
2012年第1期33-37,共5页
Journal of National University of Defense Technology
基金
国防科技大学优秀研究生创新资助项目(B080702)
关键词
光纤激光
脉冲激光
相干合成
相位控制
随机并行梯度下降算法
fiber laser
pulsed laser coherent beam combining phase control stochastic parallel gradient descent algorithm