中国激光, 2013, 40 (s1): s105008, 网络出版: 2013-12-25
基于广义回归神经网络提取布里渊谱的应变特征
Strain Characteristic Extraction of Brillouin Spectrum Based on General Regression Neural Network
光纤光学 布里渊谱 广义回归神经网络 光纤传感器 fiber optics Brillouin spectrum general regression neural network optical fiber sensor
摘要
基于光纤应变分布与布里渊散射谱频移的关系,提出了利用广义回归神经网络提取布里渊谱的应变特征方法。将布里渊谱的频率、增益分别作为广义回归神经网络的输入矢量和目标矢量,对广义回归神经网络进行训练和仿真,计算出调节权值和阈值,从而获得更加精确的布里渊谱频移。仿真实验结果和理论分析表明,与非线性最小二乘法、反向传播神经网络、径向基函数网络预测布里渊谱的应变特征相比,广义回归神经网络能够获得更精确的布里渊谱特征,相应的光纤应力误差最小,在1%之内。
Abstract
The method of extracting Brillouin spectrum characteristics by using the general regression neural network is proposed based on the relationship between the optical fiber strain and the Brillouin spectrum frequency shift. The Brillouin spectrum frequency shift and gain are taken as the input vector and target vector of general regression neural network, respectively. Then the general regression neural network is trained and simulated. The more accurate Brillouin spectrum frequency shift can be calculated with the obtained weight and threshold. Experimental results and theoretical analysis show that the Brillouin spectral feature and optical fiber strain obtained by using the general regression neural network are more accurate compared with the nonlinear least square method、back propagation neural network and radial basis function network, and the corresponding optical fiber strain error is the least (within 1%).
张志辉, 张鹏, 韩顺利. 基于广义回归神经网络提取布里渊谱的应变特征[J]. 中国激光, 2013, 40(s1): s105008. Zhang Zhihui, Zhang Peng, Han Shunli. Strain Characteristic Extraction of Brillouin Spectrum Based on General Regression Neural Network[J]. Chinese Journal of Lasers, 2013, 40(s1): s105008.