中文摘要：

提出基于模式搜索法的光纤有源自动对准算法，实现了多自由度同时寻优，解决了不同自由度之间的交叉耦合问题，提高了对准速度和对准精度。通过仿真与实验研究，与传统的爬山法进行对比。仿真结果表明，激光二极管与单模光纤五自由度对准时，模式搜索法只需20次迭代就可以找到最大点，收敛速度是爬山法的9倍。实验结果证明，横向调整两个自由度对准时，模式搜索法搜索速度比爬山法平均快10s，定位成功率达到90％。

英文摘要：

Application of the pattern search me for adjusting misalignments on multiple axial thod dire to fiber active alignment automation was developed ctions simultaneously. The interaction problem among axes was solved and the potential risk of missing the real peak was avoided, so that the alignment process time was reduced, and the alignment accuracy for locating the optimal fiber-optical coupling positioi3 was improved. The comparison of simulation and experimental results between pattern search and hill climbing method was presented. Simulation results indicate that the convergence of pattern search algorithm is nine times as fast as the hill-climbing method; it converges to the maximum coupling efficiency within 20 i ser diode and the single mode fiber. terations for the five degrees of freedom alignment between the laser diode and the single mode fiber. The experimental results also demonstrate that the average time for locating the optimal coupling position in x-y plane using pattern search algorithm is lO s faster than that of hill-climbing method; and the global convergence of the pattern search algorithm is much better than that of the hill-climbing method.