基于随机并行梯度下降算法的光束相干合成技术

Coherent beam combining experiments based on stochastic parallel gradient descent algorithm

1.
Key Laboratory of Science and Technology on High Energy Laser,CAEP,Mianyang 621900,China; 2.Institute of Applied Electronics,CAEP,P.O.Box 919-1013,Mianyang 621900,China

摘要: 介绍了随机并行梯度下降算法的基本原理，对算法流程进行了仿真验证，并对其中随机扰动幅度和增益系数两个关键参数进行了仿真分析。分析结果表明，这两个参数存在一个最适区间，只有在此区间内取值时算法才能有效收敛。以仿真分析为依据开展了光纤激光的相干合成实验，结果表明光束相干合成效果显著，有效地验证了仿真分析的结果。
- 随机并行梯度下降算法 /
- 光纤激光 /
- 相干合成 /
- 高功率激光
Abstract: The principle of stochastic parallel gradient descent (SPGD) algorithm is introduced, and the algorithm flow is verified through simulation. Two critical factors, the stochastic perturbation and the gain coefficient, are especially analyzed. The simulation results show that there is a most appropriate interval for selecting the two factors. Only with the two factors selected in this interval, the algorithm can achieve the best convergence value. Based on the simulation results, the coherent beam combining experiments are carried out with fiber lasers, resulting in significant effect of beam combining. The experimental results prove the results of simulation above. In conclusion, the research results would improve the design of coherent beam combining experiments for high power laser in the future.
- stochastic parallel gradient descent algorithm /
- fiber laser /
- coherent beam combination /
- high power laser

1.	蒋琳，刘军，袁晓蓉，张浩，李春领，蔡光明. 薄片激光器低畸变冷却技术研究. 中国激光. 2021(03): 7-13 . 百度学术
2.	郭嘉伟，贾凯，杨峰，彭春，汪晓聪，刘晓旭，韩聚洪，荣克鹏，安国斐，蔡和，王浟. 喷流冷却复合陶瓷薄片激光器的热特性仿真分析及实验研究（英文）. 光子学报. 2020(03): 7-15 . 百度学术
3.	李国会，徐宏来，吴晶，杜应磊，向汝建，周志强，于益，胡平，向振佼，张越. 非稳腔薄片激光器腔内像差组合式主动校正技术. 中国激光. 2020(10): 30-37 . 百度学术
4.	李隆，潘晓瑞，耿鹰鸽. LD端面抽运Nd:YAG微片热容激光器温度场. 激光与光电子学进展. 2017(12): 261-270 . 百度学术
5.	邵娜，朱晓，朱广志，王牧，冯玙璠，黄彦. 碟片激光器晶体模块阵列射流冲击冷却的研究. 激光技术. 2016(05): 695-700 . 百度学术