提出了一种基于激光拍频测量实现光纤光栅温度传感解调的方法,并构建了三层BP神经网络模型对温度传感数据进行优化。该方法分别采用线性啁啾光栅(CFBG)和传感光纤光栅(FBG)作为光纤激光系统的反馈腔镜,测量激光器拍频随传感光栅温度的变化实现温度传感。为减小CFBG非线性时延及时延抖动引起的温度测量误差,采用三层BP神经网络模型,将所测的拍频频率与对应温度经过BP神经网络训练。实验中,重复测量10次,得到10组拍频频率/温度数据。随机选择9组频率数据作为训练校正集,送入三层 BP 网络模型的输入层作为网络输入值,其相应的实际温度值作为网络的输出值,训练网络的权值和阈值,直至满足设定的目标使网络的参数及结构最优化。另一组作为验证样本集,测试该网络模型的实用性。此组数据的温度灵敏度和相关系数分别为37.89 kHz/℃和99.767％,对该组数据训练温度校正及预测,其相关系数达到99.95％。结果表明,利用三层BP神经网络算法对实验数据进行校正,能够有效地提高系统的测量精度。

A method of fiber Bragg grating temperature sensing demodulation based on laser beat frequency measurement is proposed, and a three layer BP neural network model is formed to optimize the temperature sensing data. In the method, the linear chirped grating (CFBG) and sensing fiber grating are respectively adopted as feedback cavity mirror of the fiber laser system. Temperature sensing is realized by measuring the change of laser beat frequency with the temperature of the sensing grating. The corresponding actual temperature value is considered as the output value of the network, the weights and thresholds of the network are trained until to meet the set target, making the parameters and structure of the network are optimal. The remaining group of data is taken as a validation sample to test the usefulness of the network model. The temperature sensitivity and correlation coefficient of this set of data are 37. 89 kHz/℃and 99. 767%, respectively, and their correlation coefficient is 99. 95%. The results show that the three layer BP neural network algorithm is utilized to correct the experimental data, which can effectively improve the measurement accuracy of the system.