Research On Online Monitoring Of Carbon Concentration Based On Laser Induced Breakdown Spectroscop

The change of global climate affects many aspects of people’s lives,for this our country also put forward the “carbon peak goal” and “carbon neutral vision” to alleviate the global greenhouse effect and to deal with the severe climate change,so carbon has become an important research object.At present,there are few researches on the spectroscopy of systematic detection of carbon concentration.Other research methods are mostly affected by some factors including equipment and background noise,which have certain limitations.For the detection of carbon concentration in atmosphere,a set of laser online detection system based on laser-induced breakdown spectroscopy(LIBS)is independently studied and established,meanwhile,combined with neural network algorithm,carbon concentration detection research and a series of studies based on carbon are carried out in this paper.The main works of this article could be summarized as follows:(1)Research and construction of online laser detection system based on LIBS,and its practical application test and feasibility verification.In view of the set of laser detection platform,the carbon in solid is selected as the research object,and the eggshell is taken as the example to carry out the practical application and feasibility verification,focusing on the comprehensiveness and accuracy of elements in the spectrum as well as the detection ability of trace elements.What’s more,principal component analysis(PCA)and neural network are used to process and further explore the data in the subsequent result analysis.Therefore,the feasibility of using PCA as the data processing method and applying neural network algorithm to the accuracy test of classification results are also verified.Eggshell is also taken as the example and finally proves that the experimental system has comprehensive and accurate spectral line detection ability and data processing method is fit for the experimental research.(2)Spectral line detection of air during candle burning.The carbon in gas is studied experimentally based on the established system.The candle combustion is taken as the example to detect the changes of air during the combustion,the spectral line of C and CN are focused and the variation of the intensity of C spectral line with time is fitted.What’s more,an appropriate way of presenting the spectral line variation result is selected.(3)The experimental testing of carbon concentration in common scenes in life.Three common scenarios with different carbon concentration are selected: air,exhaled gas and kerosene burning.Firstly,the relationship between carbon line intensity and the corresponding carbon concentration is defined by the data of the intensity and the corresponding concentration in the air and exhaled gas,and the relationship expression of the two is obtained through calculation.Then,dynamic monitoring of carbon concentration changes in human respiration and kerosene combustion is carried out to study the spectral line changes of different elements in the combustion process,in addition,the degree of combustion is predicted by the spectral line changes of carbon.(4)A scheme of carbon concentration detection in methane environment is designed,which is a combination of static detection and dynamic detection.After the methane environment is confirmed by static detection,the change of carbon concentration and the change of each element in the spectral line can be observed by the change of the concentration of methane.In this paper,an online detection system for carbon concentration in air is developed based on LIBS and neural network algorithm,which is successfully applied to the detection of carbon concentration changes during combustion and under different concentrations of methane.The development of this system realizes the real-time online monitoring of carbon concentration change,provides a new idea and method for the monitoring,and also provides an important technical support.