基于标准样品回归算法和腔增强光谱的NO_(2)检测方法

NO_(2) gas detection based on standard sample regression algorithm and cavity enhanced spectroscopy

腔增强吸收光谱技术作为一种高灵敏的痕量气体测量技术,其吸收光谱的浓度反演是极其关键的环节.为消除因吸收截面和仪器响应函数的不确定性引入的测量误差,本文提出了一种基于标准样品吸收光谱的浓度回归算法,该方法在浓度反演过程上进行优化,采用标准气体样品吸收光谱直接拟合未知浓度气体吸收光谱.采用中心波长在440 nm处的蓝色发光二极管(LED)作为光源,建立了一套非相干光腔增强吸收光谱技术(IBBCEAS)系统,实测腔镜反射率为99.915%,利用NO_(2)气体的实测吸收光谱对该算法的有效性进行了验证.与常规吸收截面回归算法比较,结果表明本文提出的标准样品回归算法具有显著的优越性,测量精度提升约4倍.利用改进的算法结合标准样品配制的多个NO_(2)气体对实验系统性能进行了深入评估,测量结果与理论值具有很好的一致性.Allan方差分析显示在积分时间为360 s的情况下,NO_(2)检测限可达到5.3 ppb(1 ppb=10^(–9)).

Cavity-enhanced absorption spectroscopy is a highly sensitive trace gas measurement technology,and the algorithm for retrieving gas concentrations is critical.The absorption cross-section is traditionally used to retrieve the concentration.However,the absorption cross-section used in the fitting process is affected not only by the response function of the instrument and the light source,but also by temperature and pressure.The uncertainty of the absorption cross-section will influence the accuracy of the result.Therefore,in order to eliminate the measurement error introduced by the uncertainty of the absorption cross-section and the instrument response function,a concentration regression algorithm based on the absorption spectrum of the standard sample is proposed.The process of concentration inversion is optimized.The absorption spectrum of standard gas is used to fit the unknown spectrum.In order to verify the effectiveness of the algorithm,the incoherent cavity enhanced absorption spectroscopy(IBBCEAS)system based on a blue light-emitting diode(LED)operating at 440 nm is established to analyze the absorption spectrum of NO_(2);and the fitting effect,measurement accuracy and stability are compared with the counter parts from the traditional absorption crosssection method.In the experiment,the measured reflectance of the cavity mirror is 99.915%.Compared with the conventional absorption cross-section regression algorithm,the standard sample regression algorithm proposed in this paper shows good results,in which the measurement accuracy is increased by about quadruple.The Allan deviation shows that a detection limit of 5.3 ppb can be achieved at an integration time of 360 s.Finally,the performance of the experimental system is evaluated by measuring the NO_(2) with different concentrations prepared by standard samples.The result shows good agreement with the theoretical value,which indicates that the improved spectral analysis algorithm is feasible and reliable for gas detection.This method can be used not only to measure NO_(2),but also to detect other gases,which shows great potential applications in monitoring the industrial emissions,atmospheric chemistry and exhaled breath analysis.