光纤光栅解调系统的寻峰算法研究

Research of peak-detection algorithm in a fiber grating demodulation system

为了提高光纤光栅波长解调系统的反射谱寻优精度,利用遗传算法的未成熟收敛性进行了研究。阐明了遗传算法提高解调精度的原理,以可调谐F-P腔的光纤光栅解调系统为基础,验证了算法的可行性和可靠性。使用MAT-LAB对F-P腔滤波光纤光栅解调系统的反射光谱进行寻峰,得到传感光纤光栅的中心波长。分析了算法中初始种群数量、遗传代数、变异率对的寻优结果影响,并得到一组最佳遗传算法参量。算法精度达到3pm数量级,50次重复计算数据标准差小于0.5pm。结果表明,该算法稳定高效、实用性强,可以显著减小噪声对光纤光栅传感系统的影响,提高解调精度。

In order to improve the precision of peak-detection of a fiber Bragg grating demodulation system, it was studied based on the premature convergence characteristics of the genetic algorithm. It was expounded how the genetic algorithm can improve the accuracy of fiber Bragg grating demodulation. Taking a fiber Bragg grating demodulation system based on fiber Fabry- Perot （F-P）wavelength filter as example, the feasibility and reliability of the algorithm were testified. The peak reflected spectrum of the fiber Fahry-Perot filter fiber Bragg grating demodulation system was detected with MATLAB and its central wavelength was obtained. The effect of initial population number, genetic algebra, mutation rate on the peak detection of the algorithm was analyzed and a set of optimal parameters were obtained. The precision was equal or less than 3pm, standard deviation was less than 0. 5pm for 50 repetition computations. The results show that the algorithm is stable, effective, and practical, can significantly reduce the noise of a fiber grating sensor system, improve the precision of demodulation.