Optimization Method Of Molten Steel Composition Detection Based On LIBS Technology

The Laser-Induced Breakdown Spectroscopy(LIBS)technique,is a typical atomic emission spectrometry element analysis technique.It has the capability of on-line quick,remote non-contact,multi-element simultaneous measurement and analysis.Compared with the traditional metallurgical detection technology,it has obvious advantages and provides a reliable scheme for on-line quick analysis and detection.In recent years,LIBS technology has penetrated into many fields,showing its outstanding application prospects and research value.In order to improve the accuracy of LIBS quantitative analysis,the key parameters and optimization algorithms of LIBS quantitative analysis were studied by the related principles and experimental conditions of LIBS.The main work completed includes:1)Deeply studied the principle of Laser-induced breakdown spectroscopy technology,summarized the requirements of modern steel composition detection in metallurgical analysis field,and built the LIBS experimental platform for on-line detection of molten steel.The principles and methods of qualitative and quantitative analysis of LIBS were reviewed and summarized in detail.2)Aiming at the selection of experimental conditions for spectral measurement,the experimental parameters such as laser energy,lens to sample distance,spectral acquisition delay and selection of effective spectra were optimized by using the spectral signal-noise ratio of Mn element as the index to ensure the reliability of spectral data.3)For the molten steel with complex chemical system,researched the four elements of Mn,Ni,Cr and Si in molten steel,proposed an optimization algorithm of support vector regression combined with catfish particle group,and established a Cat-fish PSOSVR calibration model.The experimental results showed that compared with the conventional SVR calibration model,the Cat-fish PSO-SVR calibration model has a wide range of applications and high accuracy.It provides a reference optimization algorithm for online quantitative analysis of molten steel element content in LIBS.Figure 22;Table 8;Reference 57...