OPT OpenIR  > 光谱成像技术研究室
On-line dynamic monitoring automotive exhausts: Using BP-ANN for distinguishing multi-components
Zhao, Yudi1,2; Wei, Ruyi1,2; Liu, Xuebin1,2
2017
会议名称Applied Optics and Photonics China: Optical Spectroscopy and Imaging, AOPC 2017
会议录名称AOPC 2017: Optical Spectroscopy and Imaging
卷号10461
会议日期2017-06-04
会议地点Beijing, China
出版者SPIE
产权排序1
摘要

Remote sensing-Fourier Transform infrared spectroscopy (RS-FTIR) is one of the most important technologies in atmospheric pollutant monitoring. It is very appropriate for on-line dynamic remote sensing monitoring of air pollutants, especially for the automotive exhausts. However, their absorption spectra are often seriously overlapped in the atmospheric infrared window bands, i.e. MWIR (3∼5μm). Artificial Neural Network (ANN) is an algorithm based on the theory of the biological neural network, which simplifies the partial differential equation with complex construction. For its preferable performance in nonlinear mapping and fitting, in this paper we utilize Back Propagation-Artificial Neural Network (BP-ANN) to quantitatively analyze the concentrations of four typical industrial automotive exhausts, including CO, NO, NO2and SO2. We extracted the original data of these automotive exhausts from the HITRAN database, most of which virtually overlapped, and established a mixed multi-component simulation environment. Based on Beer-Lambert Law, concentrations can be retrieved from the absorbance of spectra. Parameters including learning rate, momentum factor, the number of hidden nodes and iterations were obtained when the BP network was trained with 80 groups of input data. By improving these parameters, the network can be optimized to produce necessarily higher precision for the retrieved concentrations. This BP-ANN method proves to be an effective and promising algorithm on dealing with multi-components analysis of automotive exhausts. © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.

作者部门光谱成像技术实验室
DOI10.1117/12.2285325
收录类别EI ; ISTP
ISBN号9781510614031
语种英语
ISSN号0277786X
引用统计
被引频次:1[WOS]   [WOS记录]     [WOS相关记录]
文献类型会议论文
条目标识符http://ir.opt.ac.cn/handle/181661/29925
专题光谱成像技术研究室
作者单位1.Xi'An Institute of Optics and Precision Mechanics of Chinese Academy of Sciences, Key Laboratory of Spectral Imaging Technology of Chinese Academy of Sciences, Xi'an, 710119, China
2.University of Chinese Academy of Sciences, Beijing, 100190, China
推荐引用方式
GB/T 7714
Zhao, Yudi,Wei, Ruyi,Liu, Xuebin. On-line dynamic monitoring automotive exhausts: Using BP-ANN for distinguishing multi-components[C]:SPIE,2017.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
On-line dynamic moni(804KB)会议论文 限制开放CC BY-NC-SA请求全文
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[Zhao, Yudi]的文章
[Wei, Ruyi]的文章
[Liu, Xuebin]的文章
百度学术
百度学术中相似的文章
[Zhao, Yudi]的文章
[Wei, Ruyi]的文章
[Liu, Xuebin]的文章
必应学术
必应学术中相似的文章
[Zhao, Yudi]的文章
[Wei, Ruyi]的文章
[Liu, Xuebin]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。