Study On The Detection And Prediction Of Methane Gas Concentration Based On PC-PBGF

Air pollutants are foreign substances that enter the near-surface or lower atmosphere in gaseous form.Air pollution not only damages the survival and development of organisms,but also causes the corrosion of objects.As one of the main polluting gases,the real-time online detection of methane concentration has become an important task to protect the ecological environment.In this paper,the optical detection is the main line,and the air-core photonic crystal bandgap fiber is used as the sensing element to analyze the concentration of methane.Combined with the spectral absorption detection technology and neural network prediction technology,the methane concentration detection system is constructed.The main research work of the thesis is as follows:(1)According to the movement form and absorption characteristics of gas molecules,the absorption spectrum theory of methane gas is analyzed to determine the center wavelength corresponding to the absorption peak of methane.By comparing the advantages and disadvantages of spectral absorption differential detection and harmonic detection,the detection method of methane gas concentration was determined.(2)Construct the structural model of the experimental system,and analyze the specific theoretical knowledge of the light source part,the gas chamber part,the photoelectric conversion part,the harmonic signal extraction part,and the data acquisition part,and select the experimental equipment required for each part.In the air chamber part,a proposal was made to analyze the air-core photonic crystal band gap fiber as a sensor probe,analyze the light guiding mechanism,and determine the air-core direct and cladding clearance to design the air chamber.(3)Thirty different concentrations of methane gas are configured to absorb light intensity.Using the overall average empirical mode decomposition algorithm to perform denoising experiments on the absorption spectrum.Linear experiment for detecting methane gas concentration based on spectral absorption harmonics.(4)Analyze the theoretical basis of the BP neural network and the optimized PSO-BP neural network,and apply it to the quantitative analysis experiment of methane gas concentration to achieve the effect of predicting the concentration.