Applications of optical measurement technology in pollution gas monitoring at thermal power plants

2011 
This paper presents the work of using advanced optical measurement techniques to implement stack gas emission monitoring and process control. A system is designed to conduct online measurement of SO 2 /NO X and mercury emission from stacks and slipping NH 3 of de-nitrification process. The system is consisted of SO 2 /NOX monitoring subsystem, mercury monitoring subsystem, and NH 3 monitoring subsystem. The SO 2 /NO X monitoring subsystem is developed based on the ultraviolet differential optical absorption spectroscopy (UV-DOAS) technique. By using this technique, a linearity error less than ±1% F.S. is achieved, and the measurement errors resulting from optical path contamination and light fluctuation are removed. Moreover, this subsystem employs in situ extraction and hot-wet line sampling technique to significantly reduce SO 2 loss due to condensation and protect gas pipeline from corrosion. The mercury monitoring subsystem is used to measure the concentration of element mercury (Hg 0 ), oxidized mercury (Hg 2+ ), and total gaseous mercury (Hg T ) in the flue gas exhaust. The measurement of Hg with a low detection limit (0.1μg/m 3 ) and a high sensitivity is realized by using cold vapor atom fluorescence spectroscopy (CVAFS) technique. This subsystem is also equipped with an inertial separation type sampling technique to prevent gas pipeline from being clogged and to reduce speciation mercury measurement error. The NH 3 monitoring subsystem is developed to measure the concentration of slipping NH 3 and then to help improving the efficiency of de-nitrification. The NH 3 concentration as low as 0.1ppm is able to be measured by using the off-axis integrated cavity output spectroscopy (ICOS) and the tunable diode laser absorption spectroscopy (TDLAS) techniques. The problem of trace NH 3 sampling loss is solved by applying heating the gas pipelines when the measurement is running.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    1
    Citations
    NaN
    KQI
    []