Rapid detection of trace sulfur content in ship fuel oil based on tin oxide quantum dot fluorescent sensors assisted by multi-column convolutional neural network.
Saved in:
| Title: | Rapid detection of trace sulfur content in ship fuel oil based on tin oxide quantum dot fluorescent sensors assisted by multi-column convolutional neural network. |
|---|---|
| Authors: | Fu, Ce1,2 (AUTHOR), Li, Hongjin1 (AUTHOR), Li, Wenping1 (AUTHOR), Ding, Chenwen1 (AUTHOR), Zhang, Yanan1 (AUTHOR), Zhai, Zhaoxia1 (AUTHOR), Liu, Jianqiao1,2 (AUTHOR) jqliu@dlmu.edu.cn, Wang, Junsheng1,2 (AUTHOR) wangjsh@dlmu.edu.cn |
| Source: | Microchemical Journal. Oct2024, Vol. 205, pN.PAG-N.PAG. 1p. |
| Subjects: | Convolutional neural networks, Generative adversarial networks, Stannic oxide, Fluorescence spectroscopy, Data augmentation |
| Abstract: | [Display omitted] • The method accomplishes a classification algorithm for fluorescence spectra of SnO 2 QDs. • The method contributes to the combination of machine learning and nanotechnology. • The use of MCNN improves the classification performances of the classifier. • The data augmentation method based on WGAN provides a solution for the small sample size problem. The pollutant emission from ships contributes a significant part of the total emission from the transportation industry. Sulfur dioxide (SO 2), generated by the combustion of marine heavy fuel oil with high sulfur content, is one of the most common gaseous contaminants in ship exhaust. Thus, it is necessary to develop a fast technique to determine the fuel sulfur content (FSC) for the monitor and control of fuel quality in ships. In this paper, a new FSC prediction model is proposed for fluorescence emission spectra of tin oxide quantum dots (SnO 2 QDs). The indistinguishable features of fluorescence spectra are determined by the multi-column convolutional neural network (MCNN) classifier that incorporate transfer learning and the Wasserstein generative adversarial network (WGAN). The prediction of FSC values is enabled by the classification algorithm, which extracts all the features from the fluorescence spectrum for each fuel catalog. The predicting accuracy of the present model is over 97.14 %, demonstrating the potential applicability of this fast FSC detection technique. This work not only accomplishes a classification algorithm for fluorescence spectra of SnO 2 QDs, but also contributes to the combination of machine learning and nanotechnology. [ABSTRACT FROM AUTHOR] |
| Copyright of Microchemical Journal is the property of Elsevier B.V. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
Be the first to leave a comment!
