Font Size: a A A

Quantitative Analysis Study Of Fiber-Optic Laser–Induced Breakdown Spectroscopy

Posted on:2020-02-13Degree:MasterType:Thesis
Country:ChinaCandidate:W J LuFull Text:PDF
GTID:2370330590458302Subject:Optical Engineering
Abstract/Summary:PDF Full Text Request
Laser probe,also known as Laser-induced breakdown spectroscopy(LIBS),has been preliminarily used in environmental protection,space exploration,biomedical and other fields because of its in-situ,real-time,and all-element advantages.However,traditional LIBS devices are complex and poorly portable,and cannot adapt to complex industrial monitoring requirements.Since fiber-optic laser-induced breakdown(FO-LIBS)has the advantages of simple device,good portability,and ability to adapt to complex detection environments,FOLIBS has attracted the attention of researchers in the field of industrial detection such as nuclear reactor detection and deep water substance testing.However,due to the fiber damage threshold,the single pulse energy transmitted by the fiber is limited,and the quantitative accuracy is poor.The quantitative accuracy and sensitivity of FO-LIBS is much larger than that of traditional LIBS.Therefore,this paper proposed two solutions of multivariate data analysis and spatial resolution acquisition.which can improve the self-absorption effect,the accuracy and sensitivity of quantitative analysis effectively.The main achievements and innovations are as follows:(1)To solve the problem that the univariate analysis method can not make full use of the spectral information,leading to the low precision of FO-LIBS quantitative analysis,a multivariate analysis method is proposed.In this paper,four multivariate analysis methods(PLS,SPLS,SVM and LS-SVM)are applied.SPLS has built-in variable selection based on the PLS model,which can eliminate the irrelevant variables in the spectrum and improve the accuracy of quantitative analysis.Since both SVM and LS-SVM are linear models and cannot fit the nonlinear problems existing in the spectrum,the nonlinear relationship in the spectrum was taken to promote the accuracy.The training calculation of LS-SVM is simple,and the prediction accuracy of SVM is higher.Finally,the SVM model is used to analyze the Mn,Cr,Ni and Ti elements in pig iron samples.The obtained RMSEPs of Mn,Cr,Ni and Ti in pig iron samples are 0.0982,0.0185,0.0179,0.0178 wt.%,and AREs are 17.17%,17.05%,24.53% and 26.57%,respectively.(2)The spatially resolved method was first used in FO-LIBS.By observing the spatial distribution difference of plasma spectra in aluminum alloy samples,it was found that FOLIBS induced plasma has weak intensity at the center and strong intensity at the edge.By selecting the appropriate acquisition position in the X-axis direction,the influence of the selfabsorption effect on the spectrum can be overcome,and the quantitative analysis accuracy can be improved.Finally,the RMSECV of Fe,Mg and Zn decreased from 0.388,0.348,0.097 wt.% to 0.217,0.172,0.024 wt.%,respectively;and the detection limits of the Fe,Mg and Zn elements were 502.07,325.83,and 304.05 ug/g,respectively.In summary,the multivariate data analysis method and spatial resolution acquisition method studied in this paper can effectively overcome the influence of weak spectral intensity and strong self-absorption effect on the quantitative analysis performance of fiber-optic energy-transmitting laser probe technology.These two methods can effectively improve quantitative accuracy and sensitivity of FO-LIBS and provide theoretical and technical support for the practical application of FO-LIBS.
Keywords/Search Tags:laser-induced breakdown spectroscopy, fiber-optic, spatial resolution, self-absorption, data analysis
PDF Full Text Request
Related items