摘要
传统差分吸收光谱(differential optical absorption spectroscopy,DOAS)技术对长光程或高浓度烟气污染物可以实现准确、快速和在线测量,但是对于短光程、低浓度污染气体测量却存在较大的浓度反演误差。该文提出了一种新的光谱数据处理方法,并对信噪比较低的实测光谱数据进行了浓度反演。研究结果表明:气体浓度在0～150mg/m3范围内,新算法的反演精度明显优于传统算法,反演误差都在10%以内;在150～4000mg/m3范围内与传统算法相当;超过4000mg/m3后新算法的反演误差较大。新算法具有良好的抗烟尘干扰能力,其零点误差较传统算法减小了95%以上。与传统DOAS算法相结合可以获得较宽的动态测量范围。
Traditional differential optical absorption spectroscopy （DOAS） technique can be applied to realize accurate, rapid and in-situ concentration measurement to flue gas of long light path or high concentration while large error will occur when applied to flue gas measurement of low concentration at short light path. A new processing method of spectral data was put forward and applied to the concentration inversion of experimental spectral data at low signal to noise ratio （SNR）. Research results show： innovative algorithm is obviously better than traditional one with the less than 10% error of inversion of the concentration in the range of 1.4- 150mg/m^3, and agree with traditional algorithm in the range of 150-4 000 mg/m^3. But the error of innovative algorithm is larger than traditional one when the concentration is over 4 000 mg/m^3. It has a good performance of anti-jamming from dust to low concentration gas at short light path, and the zero point error is reduced over 95% than the traditional DOAS. The broad dynamic measurement range can be achieved combined with traditional DOAS.
出处
《中国电机工程学报》
EI
CSCD
北大核心
2008年第23期65-70,共6页
Proceedings of the CSEE
基金
国家自然科学基金项目(50576010)
教育部博士点基金资助项目(20040286038)~~
关键词
差分吸收光谱法
最小二乘算法
短光程
低浓度
污染物
浓度反演
differential optical absorption spectroscopy
least-squared method
short light path
low concentration
pollutants
concentration inversion
