期刊
SPECTROCHIMICA ACTA PART B-ATOMIC SPECTROSCOPY
卷 113, 期 -, 页码 167-173出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.sab.2015.09.021
关键词
Laser-induced breakdown spectroscopy (LIBS); Coal quality analysis; Support vector regression (SVR); Principal component analysis (PCA); Pulsed laser energy stabilization
类别
资金
- National Natural Science Foundation of China [61475093, 61127017, 61378047, 61205216, 61178009, 61108030, 61275213]
- National Key Technology RD Program [2013BAC14B01]
- 973 program [2012CB921603]
- Shanxi Natural Science Foundation [2013021004-1]
- Shanxi Scholarship Council of China [2013-011, 2013-01]
- Program for the Outstanding Innovative Teams of Higher Learning Institutions of Shanxi [20121401120016]
- State Key Laboratory of Power System [SKLD15KZ07]
It is vitally important for a power plant to determine the coal property rapidly to optimize the combustion process. In this work, a fully software-controlled laser-induced breakdown spectroscopy (LIBS) based coal quality analyzer comprising a LIBS apparatus, a sampling equipment, and a control module, has been designed for possible application to power plants for offering rapid and precise coal quality analysis results. A closed-loop feedback pulsed laser energy stabilization technology is proposed to stabilize the Nd: YAG laser output energy to a preset interval by using the detected laser energy signal so as to enhance the measurement stability and applied in a month-long monitoring experiment. The results show that the laser energy stability has been greatly reduced from +/- 5.2% to +/- 13%. In order to indicate the complex relationship between the concentrations of the analyte of interest and the corresponding plasma spectra, the support vector regression (SVR) is employed as a non-linear regression method. It is shown that this SVR method combined with principal component analysis (PCA) enables a significant improvement in cross-validation accuracy by using the calibration set of coal samples. The root mean square error for prediction of ash content, volatile matter content, and calorific value decreases from 2.74% to 1.82%, 1.69% to 1.22%, and 123 MJ/kg to 0.85 MJ/kg, respectively. Meanwhile, the corresponding average relative error of the predicted samples is reduced from 83% to 5.48%, 5.83% to 4.42%, and 5.4% to 3.68%, respectively. The enhanced levels of accuracy obtained with the SVR combined with PCA based calibration models open up avenues for prospective prediction in coal properties. (C) 2015 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据