Application Progress of Artificial Neural Network in Laser-Induced Breakdown Spectral Data Analysis

Wen-ya ZHAO; Hong MIN; Shu LIU; Ya-rui AN; Jin YU

doi:10.3964/j.issn.1000-0593(2021)07-1998-07

Journals >Spectroscopy and Spectral Analysis >Volume 41 >Issue 7 >Page 1998 > Article

Spectroscopy and Spectral Analysis
Vol. 41, Issue 7, 1998 (2021)

Application Progress of Artificial Neural Network in Laser-Induced Breakdown Spectral Data Analysis

Wen-ya ZHAO^1、*, Hong MIN^2、2;, Shu LIU^{2、2; *;}, Ya-rui AN^{1、1; *;}, and Jin YU^3、3;

Author Affiliations

¹1. College of Science, University of Shanghai for Science and Technology, Shanghai 200093, China

²2. Technical Center for Industrial Product and Raw Material Inspection and Testing, Shanghai Customs, Shanghai 200135, China

³3. School of Physics and Astronomy, Shanghai Jiao Tong University, Shanghai 200240, China

show less

DOI: 10.3964/j.issn.1000-0593(2021)07-1998-07 Cite this Article

Wen-ya ZHAO, Hong MIN, Shu LIU, Ya-rui AN, Jin YU. Application Progress of Artificial Neural Network in Laser-Induced Breakdown Spectral Data Analysis[J]. Spectroscopy and Spectral Analysis, 2021, 41(7): 1998 Copy Citation Text

show less

Fig. 1. The composition of LIBS instrument^[8]

Download full size | View in the Article

Fig. 2. Typical flow chart for processing mineral LIBS spectral data^[13]

Download full size | View in the Article

Fig. 3. Fs-LIBS combined with ANN flow chart for plastic classification^[27]

Download full size | View in the Article

Fig. 4. Flow chart of LIBS data analysis for identifying biomass pellets^[37]

Download full size | View in the Article

合金类型	元素	ANN模型	主要成果	参考文献
碳钢	C(Ⅰ) 247.86 nm	GA-BPANN	克服相邻铁谱线干扰	[20]
碳钢; 低合金钢; 微合金钢	Cu; V	GA-ANN	GA用于选取样本目标元素的特征谱线强度比, 提高了ANN分析精度	[21]
碳钢; 低合金钢; 微合金钢	Cr; Ni	基于ANN的多谱线校正(MSLC)	用目标和基体元素的多谱线强度比训练神经网络识别等离子体脉冲的变化, 克服了激发条件不稳定的问题和自吸收效应	[22]
铜镍二元合金	Ni	BPANN	训练ANN识别谱线强度与等离子体参数之间的基本物理关系, 克服了激光波动和基体效应	[23]
青铜标样	Cu (324.75 nm; 327.39 nm) Sn (303.41 nm; 326.23 nm)	ELM-SVR	以ELM的输出为支持向量回归(SVR)输入的混合模型, 解决ELM模型的超参数过拟合的问题, 提高了定量分析的准确度	[24]

Table 1. The typical study on the analysis of alloying elements by LIBS technique combined with ANN model

View in the Article

元素	ANN模型	输入变量	主要成果	参考文献
K	CNN	主成分和延迟时间确定的时间分辨LIBS数据矩阵	R_V=0.996 8; RMSE_V=0.078 5	[32]
Cd	BPANN	分析谱线与谱线区间: (356. 29~346. 93 nm)	R_C=0.999 9; R_P=0.981 5	[33]
Pb, Cd	BPANN	分析谱线与谱线区间: (346. 29~346. 93 nm); (405.43~406.17 nm)	R_P=0.995 3; RMSE_P=0.145 2
Ag	BPANN	数据最后一列加入一项带有土壤类型的数据, 组成广义光谱	R_V=0.999 9; ppm量级	[34]

Table 2. Typical studies of LIBS technology combined with ANN model to detect soil metal content

Download Citation

Save the article for my favorites

Paper Information