基于空心激光捕获- libs和机器学习的单个微米级煤气溶胶粒子定量分析

IF 5.6 1区化学 Q1 CHEMISTRY, ANALYTICAL

Talanta Pub Date : 2025-07-11 DOI:10.1016/j.talanta.2025.128565

Jiaqiang Du , Tianlong Zhang , Hua Li , Min Wang , Tianguo Wang , Fengguang Li , Tingting Chen

{"title":"基于空心激光捕获- libs和机器学习的单个微米级煤气溶胶粒子定量分析","authors":"Jiaqiang Du , Tianlong Zhang , Hua Li , Min Wang , Tianguo Wang , Fengguang Li , Tingting Chen","doi":"10.1016/j.talanta.2025.128565","DOIUrl":null,"url":null,"abstract":"<div><div>Fine particulate matter is one of the major pollutants emitted from coal combustion. Particulate matter and gaseous pollutants emitted from coal combustion are one of the main pollutants that are concerned. Laser-induced breakdown spectroscopy (LIBS) is a powerful and versatile analytical tool for real-time, simultaneous, and all-element detection. However, its broader applications are constrained by spectral interference, matrix effects, weak spectral intensity and so on. To address this challenge, multi-element quantitative analysis methods based on hollow laser trapping assisted LIBS signal enhancement of fiber collimated system were established by machine learning. The influence of five different spectral preprocessing methods and three different variable selection methods on the prediction performance of the RF calibration model was investigated. The Savitzky-Golay convolution derivative-variable importance projection-random forest (SG-VIP-RF) (Fe) and first-order derivative-variable importance measurement-random forest (D1st-VIM-RF) (Ca) calibration models were constructed based on the optimal spectral preprocessing method and variable selection method. The prediction performance of Fe and Ca elements are shown as follows: Fe (R<sub>p</sub><sup>2</sup> = 0.9861, MREP = 0.0477, RMSEP = 1.408 %) and Ca (R<sub>p</sub><sup>2</sup> = 0.9580, MREP = 0.0627, RMSEP = 2.1461 %), and their relative standard deviation (RSD) values are 2.2 % and 5.2 %, respectively. The results demonstrate that the RF calibration model based on the optical fiber collimated LIBS signal enhancement method is successfully applied to the quantitative analysis of standard coal samples. It is expected to provide theoretical basis and technical support for in-situ online rapid monitoring of coal-fired energy materials, and further promote the wide application of LIBS in on-line monitoring fields such as environmental monitoring, geological exploration and metallurgical analysis, adhere to green and low-carbon sustainable development, and help promote the goal of carbon peak carbon neutralization.</div></div>","PeriodicalId":435,"journal":{"name":"Talanta","volume":"297 ","pages":"Article 128565"},"PeriodicalIF":5.6000,"publicationDate":"2025-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Individual micron-sized coal aerosol particle for quantitative analysis based on hollow laser trapping-LIBS and machine learning\",\"authors\":\"Jiaqiang Du , Tianlong Zhang , Hua Li , Min Wang , Tianguo Wang , Fengguang Li , Tingting Chen\",\"doi\":\"10.1016/j.talanta.2025.128565\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>Fine particulate matter is one of the major pollutants emitted from coal combustion. Particulate matter and gaseous pollutants emitted from coal combustion are one of the main pollutants that are concerned. Laser-induced breakdown spectroscopy (LIBS) is a powerful and versatile analytical tool for real-time, simultaneous, and all-element detection. However, its broader applications are constrained by spectral interference, matrix effects, weak spectral intensity and so on. To address this challenge, multi-element quantitative analysis methods based on hollow laser trapping assisted LIBS signal enhancement of fiber collimated system were established by machine learning. The influence of five different spectral preprocessing methods and three different variable selection methods on the prediction performance of the RF calibration model was investigated. The Savitzky-Golay convolution derivative-variable importance projection-random forest (SG-VIP-RF) (Fe) and first-order derivative-variable importance measurement-random forest (D1st-VIM-RF) (Ca) calibration models were constructed based on the optimal spectral preprocessing method and variable selection method. The prediction performance of Fe and Ca elements are shown as follows: Fe (R<sub>p</sub><sup>2</sup> = 0.9861, MREP = 0.0477, RMSEP = 1.408 %) and Ca (R<sub>p</sub><sup>2</sup> = 0.9580, MREP = 0.0627, RMSEP = 2.1461 %), and their relative standard deviation (RSD) values are 2.2 % and 5.2 %, respectively. The results demonstrate that the RF calibration model based on the optical fiber collimated LIBS signal enhancement method is successfully applied to the quantitative analysis of standard coal samples. It is expected to provide theoretical basis and technical support for in-situ online rapid monitoring of coal-fired energy materials, and further promote the wide application of LIBS in on-line monitoring fields such as environmental monitoring, geological exploration and metallurgical analysis, adhere to green and low-carbon sustainable development, and help promote the goal of carbon peak carbon neutralization.</div></div>\",\"PeriodicalId\":435,\"journal\":{\"name\":\"Talanta\",\"volume\":\"297 \",\"pages\":\"Article 128565\"},\"PeriodicalIF\":5.6000,\"publicationDate\":\"2025-07-11\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Talanta\",\"FirstCategoryId\":\"92\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0039914025010550\",\"RegionNum\":1,\"RegionCategory\":\"化学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"CHEMISTRY, ANALYTICAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Talanta","FirstCategoryId":"92","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0039914025010550","RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CHEMISTRY, ANALYTICAL","Score":null,"Total":0}

引用次数: 0

摘要

细颗粒物是燃煤排放的主要污染物之一。燃煤排放的颗粒物和气态污染物是人们关注的主要污染物之一。激光诱导击穿光谱（LIBS）是一种功能强大的多功能分析工具，用于实时，同时和全元素检测。然而，它的广泛应用受到光谱干扰、矩阵效应、弱光谱强度等因素的制约。为了解决这一问题，利用机器学习技术建立了基于空心激光俘获辅助LIBS信号增强的光纤准直系统多元素定量分析方法。研究了5种不同的光谱预处理方法和3种不同的变量选择方法对射频校准模型预测性能的影响。基于最优光谱预处理方法和变量选择方法，构建了Savitzky-Golay卷积导数-变重要度投影-随机森林（SG-VIP-RF）（Fe）和一阶导数-变重要度测量-随机森林（d1 - vim - rf）（Ca）标定模型。Fe和Ca元素的预测性能分别为：Fe （Rp2 = 0.9861, MREP = 0.0477, RMSEP = 1.408%）和Ca (Rp2 = 0.9580, MREP = 0.0627, RMSEP = 2.1461%)，相对标准偏差（RSD）值分别为2.2%和5.2%。结果表明，基于光纤准直LIBS信号增强方法的射频校准模型成功地应用于标准煤样品的定量分析。有望为燃煤能源材料原位在线快速监测提供理论基础和技术支撑，进一步推动LIBS在环境监测、地质勘探、冶金分析等在线监测领域的广泛应用，坚持绿色低碳可持续发展，助力推进碳峰碳中和目标的实现。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

Individual micron-sized coal aerosol particle for quantitative analysis based on hollow laser trapping-LIBS and machine learning

查看原文本刊更多论文

Individual micron-sized coal aerosol particle for quantitative analysis based on hollow laser trapping-LIBS and machine learning

Fine particulate matter is one of the major pollutants emitted from coal combustion. Particulate matter and gaseous pollutants emitted from coal combustion are one of the main pollutants that are concerned. Laser-induced breakdown spectroscopy (LIBS) is a powerful and versatile analytical tool for real-time, simultaneous, and all-element detection. However, its broader applications are constrained by spectral interference, matrix effects, weak spectral intensity and so on. To address this challenge, multi-element quantitative analysis methods based on hollow laser trapping assisted LIBS signal enhancement of fiber collimated system were established by machine learning. The influence of five different spectral preprocessing methods and three different variable selection methods on the prediction performance of the RF calibration model was investigated. The Savitzky-Golay convolution derivative-variable importance projection-random forest (SG-VIP-RF) (Fe) and first-order derivative-variable importance measurement-random forest (D1st-VIM-RF) (Ca) calibration models were constructed based on the optimal spectral preprocessing method and variable selection method. The prediction performance of Fe and Ca elements are shown as follows: Fe (R_p² = 0.9861, MREP = 0.0477, RMSEP = 1.408 %) and Ca (R_p² = 0.9580, MREP = 0.0627, RMSEP = 2.1461 %), and their relative standard deviation (RSD) values are 2.2 % and 5.2 %, respectively. The results demonstrate that the RF calibration model based on the optical fiber collimated LIBS signal enhancement method is successfully applied to the quantitative analysis of standard coal samples. It is expected to provide theoretical basis and technical support for in-situ online rapid monitoring of coal-fired energy materials, and further promote the wide application of LIBS in on-line monitoring fields such as environmental monitoring, geological exploration and metallurgical analysis, adhere to green and low-carbon sustainable development, and help promote the goal of carbon peak carbon neutralization.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Talanta 化学-分析化学

CiteScore

12.30

自引率

4.90%

发文量

861

审稿时长

29 days

期刊介绍： Talanta provides a forum for the publication of original research papers, short communications, and critical reviews in all branches of pure and applied analytical chemistry. Papers are evaluated based on established guidelines, including the fundamental nature of the study, scientific novelty, substantial improvement or advantage over existing technology or methods, and demonstrated analytical applicability. Original research papers on fundamental studies, and on novel sensor and instrumentation developments, are encouraged. Novel or improved applications in areas such as clinical and biological chemistry, environmental analysis, geochemistry, materials science and engineering, and analytical platforms for omics development are welcome. Analytical performance of methods should be determined, including interference and matrix effects, and methods should be validated by comparison with a standard method, or analysis of a certified reference material. Simple spiking recoveries may not be sufficient. The developed method should especially comprise information on selectivity, sensitivity, detection limits, accuracy, and reliability. However, applying official validation or robustness studies to a routine method or technique does not necessarily constitute novelty. Proper statistical treatment of the data should be provided. Relevant literature should be cited, including related publications by the authors, and authors should discuss how their proposed methodology compares with previously reported methods.