Design of Smoke Detection Algorithm Based on UV Spectroscopy

The difficulty of using ultraviolet spectroscopy to measure the harmful components of flue gas is that the gases have mutual interference and their characteristic absorption peaks also interfere with each other. In this paper, we propose a cyclic iterative algorithm to solve the problem of mutual interference between gases by using Lambert-Beer’s law and absorbance. The algorithm uses different UV-light wavelengths at 190 nm–290 nm for different characteristics of UV light with different absorption peaks. The iteration is repeated until the concentration difference between adjacent two gases is less than a certain value. It is considered that the elemental gas The exact concentration, and through the programming to achieve the results. The experimental results show that the algorithm can calculate the actual exact concentration of various simple gases in the mixed gas, and the accuracy error does not exceed 3%. The actual gas detection requirements can be met.