马翠红,赵士超

• 1:华北理工大学电气工程学院

摘要(Abstract)：

遗传神经网络与激光诱导击穿光谱技术(LIBS)相结合的方法能够更好地对钢液成分进行定量分析检测。建立基于遗传算法为核心的三层误差反向传播(BP)分析模型,由于BP网络的初始权值和阈值是随机数,因此存在收敛速度慢、不能保证收敛全局最优解等缺点,而遗传算法能够优化出最佳的初始权值和阈值,可以较好地克服这些问题。网络的输入选取几种元素的峰值强度与Fe元素的峰值强度进行峰值归一化处理;网络的输出为元素浓度。构建遗传神经网络定量分析模型对钢液中的Mn元素进行定量分析,得到相对标准差(RSD)为7.46%,相关系数为0.996。实验结果表明,遗传神经网络结合LIBS技术相比传统LIBS定标分析法检测的结果精确度有了一定提高。

关键词(KeyWords)： 光谱学;激光诱导击穿光谱技术;实验装置;神经网络;遗传算法;定量分析

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金(61271402)~~

作者(Author): 马翠红,赵士超

DOI: 10.16652/j.issn.1004-373x.2018.15.038

参考文献(References)：

• [1]RAI A K,YUEH F Y,SINGH J P.High temperature fiber optic laser-induced breakdown spectroscopy sensor for analysis of molten alloy constituents[J].Review of scientific instruments,2002,73(10):3589-3599.
• [2]RAI A K,YUEH F Y,SINGH J P.Laser-induced breakdown spectroscopy of molten aluminum alloy[J].Applied spectroscopy,2003,42(12):2078-2084.
• [3]FERREIRA E C,MILORI D M B P,FERREIRA E J,et al.Artificial neural networks for Cu quantitative determination in soil using a portable laser induced breakdown spectroscopy system[J].Spectrochimica acta:part B atomic spectroscopy,2008,63(10):1216-1220.
• [4]沈沁梅,周卫东,李科学.基于遗传神经网络的激光诱导击穿光谱元素定量分析技术[J].中国激光,2011,38(3):247-252.SHEN Qinmei,ZHOU Weidong,LI Kexue.Quantitative elemental analysis using laser induced breakdown spectroscopy and neuro-genetic approach[J].Chinese journal of lasers,2011,38(3):247-252.
• [5]刘锦伟,谢雄刚,方井.基于遗传算法:BP神经网络的煤层注水效果分析[J].工矿自动化,2016,42(1):48-51.LIU Jinwei,XIE Xionggang,FANG Jing.Effect analysis of coal seam water infusion based on genetic algorithm-BP neural network[J].Industry and mine automation,2016,42(1):48-51.
• [6]马翠红,王维国.基于遗传算法的LIBS钢液定量分析[J].中国冶金,2016,26(11):30-33.MA Cuihong,WANG Weiguo.Genetic algorithm liquid steel quantitative analysis based on LIBS[J].China metallurgy,2016,26(11):30-33.
• [7]李俊香,杨友良,孟凡伟,等.用于LIBS钢液在线定量分析的基体校正方法[J].激光与光电子学进展,2013,50(3):134-138.LI Junxiang,YANG Youliang,MENG Fanwei,et al.Matrix correction method used for liquid steel online quantitative analysis by LIBS[J].Laser optoelectronics progress,2013,50(3):134-138.
• [8]杨友良,王鹏,马翠红.基于粒子群优化支持向量机的LIBS钢液Mn元素定量分析[J].激光与光电子学进展,2015,52(7):289-294.YANG Youliang,WANG Peng,MA Cuihong.Quantitative analysis of Mn element in liquid steel by LIBS based on particle swarm optimized support vector machine[J].Laser optoelectronics progress,2015,52(2):289-294.
• [9]李军梅,胡以华,陶小红.基于主成分分析与BP神经网络的识别方法研究[J].红外与激光工程,2005,34(6):719-723.LI Junmei,HU Yihua,TAO Xiaohong.Recognition method based on principal component analysis and back propagation neural network[J].Infrared and laser engineering,2005,34(6):719-723.
• [10]OH S Y,YUEH F Y,SINGH J P.Quantitative analysis of tin alloy combined with artificial neural network prediction[J].Applied optics,2010,49(13):C36-C41.
• [11]汪家升,乔东坡,陆运章,等.强度归一化法定量分析岩石激光诱导击穿光谱[J].中国激光,2010,37(1):225-230.WANG Jiasheng,QIAO Dongpo,LU Yunzhang,et al.Quantitative analysis of laser-induced breakdown spectroscopy in rocks by using intensity normalization method[J].Chinese journal of lasers,2010,37(1):225-230.
• [12]李敏,朱心勇,徐媛,等.应用LIBS技术定量检测湖水样品中的铜[J].激光与光电子学进展,2013,50(1):195-198.LI Min,ZHU Xinyong,XU Yuan,et al.Quantitative determination of Cu in lake water by laser induced breakdown spectroscopy[J].Laser optoelectronics progress,2013,50(1):195-198.