Incoherent broadband cavity enhanced absorption spectrum fitting waveband selection method

Abstract

The invention discloses an incoherent broadband cavity enhanced absorption spectrum fitting waveband selection method. A BP neural network machine learning method is used for determining a nonlinear relationship between two parameters of a width and a central wavelength of a fitting wave band and relative fitting uncertainty and a fitting residual standard deviation and then a fitting waveband isdetermined according to expected relative fitting uncertainty and fitting residual standard deviation. The method specifically comprises the following steps: determining a wide spectral waveband; dividing the wide spectral band into a plurality of sub-bands with different widths and central wavelengths; fitting the sub-band absorption spectrum; normalizing a fitting result to form a data sample; establishing a BP neural network, and performing network learning; and obtaining the width and the central wavelength of a to-be-selected waveband by using the learned BP neural network, and finally obtaining a spectrum fitting waveband. Compared with the prior art, the method has the advantages that the optimal spectral fitting waveband can be obtained, and the possibility that the fitting resultof the existing method has large deviation is avoided.

Description

technical field [0001] The invention relates to a method for selecting a fitting band of a broadband absorption spectrum, in particular to a method for selecting a fitting band of an incoherent broadband cavity-enhanced absorption spectrum. Background technique [0002] Incoherent broadband cavity-enhanced absorption spectroscopy technology is a high-sensitivity optical detection method. It uses an optical resonant cavity composed of high-reflectivity mirrors to increase the absorption optical path, thereby improving the detection sensitivity of the measured gas. This technology selects the absorption spectrum of a certain band, and uses the least square method to fit the gas absorption cross section to the measured absorption coefficient to obtain the molecular number concentration of the measured gas. When selecting the spectral fitting band, it is generally necessary to consider the band where the reflectance of the lens is, the band where the radiation spectrum of the li...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problem that the existing spectral fitting band selection method of the incoherent broadband cavity enhanced absorption spectroscopy technology may produce large deviations in the fitting results of the measured gas concentration, and invented a machine learning method based on BP neural network Band selection method for spectral fitting

Images