用于分布式光纤传感系统的混合算法研究

Research on mix algorithm applied in distributed optical fiber sensor

在基于马赫-曾德干涉(MZI)原理的分布式光纤扰动传感系统中,为了减小由于偏振态退化引起的定位误差,提出了一种基于模拟退火混沌粒子群(SCPSO)算法的偏振控制算法,阐述了算法的优化过程。理论分析和实验结果表明,相对于优化前的智能优化方法,优化后的方法提高了收敛效率和收敛精度,并有效提高了系统定位精度,使得系统定位误差控制在±10m以内,完成优化所需迭代次数小于9次。

To suppress the positioning errors induced by the polarization fading in distributed optical fiber perturbation sensor based on dual Maeh-Zehnder interferometers,the optimization algorithm used in the polarization control of the distributed optical fiber perturbation sensor is studied,the inapplicable part in the particle swarm algorithm is improved,which affects the polarization adjustment of the distributed op- tical fiber perturbation sensor, and a polarization control algorithm based on simulated annealing chaotic particle swarm optimization （SCPSO） is proposed. The design concept and implementation process of the algorithm are introduced. Then the algorithm is analyzed theoretically and evaluated experimentally. The experiments are conducted in the mountainous areas where the natural environment is relatively bad. To adjust the polarization state of the distributed optical fiber perturbation sensor using this mix algorithm in the way of feedback control, the difference between the two interference signals is made to be the a- daptive value of the polarization control algorithm. Theory analysis and experiment results show that the mix algorithm improves the convergence accuracy and convergence efficiency relative to basic intelligent optimization methods,and improves the positioning accuracy dfectively. The iteration times are less than 9 to coplete the optimizaiton, and the positioning error of the distributed optical fiber perturbation sensor can be restricted within ±10m.