基于量子粒子群优化算法的光纤光栅参数重构

Reconstruction of Fiber Grating Parameters from Reflectivity Using Quantum Particle Swarm Optimization Algorithm

提出了一种基于量子粒子群优化算法的光纤光栅参数重构方法。该方法通过传输矩阵法得到优化目标函数,并将待优化的光纤光栅参数以粒子表示,再让粒子在解空间模拟量子行为进行搜索。以均匀布拉格光栅和线性啁啾光纤光栅为例,分别采用遗传算法(GA)、经典粒子群优化(PSO)算法以及量子粒子群优化(QPSO)算法对其进行参数重构。与传统粒子群算法及遗传算法相比,该方法借鉴了量子行为,具有更好的收敛性能和稳态性能。数值结果表明,种群规模为40时,针对均匀和非均匀光栅分别进化100代和200代得到的重构参数误差均小于0.5%。

A parameter reconstruction method for the physical parameters of fiber Bragg grating（FBG） based on quantum particle swarm optimization（QPSO） is proposed.In the proposed method,the objective function is constructed according to the transfer matrix theory,the physical parameters of fiber gratings are represented in the form of particle,and the optimized parameters are obtained by the particles′ searching in the solution space according to the quantum behavior.When compared with genetic algorithm（GA） and particle swarm optimization（PSO）,the proposed QPSO-based method simulates the quantum behavior,which leads to a better convergence performance and a better static-state performance.The simulation results show that,for both uniform and nonuniform fiber grating evolving 100 or 200 times,the proposed method has the reconstruction parameter error of less than 0.5% when the swarm population is 40.