We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
A stacked extreme learning machines model on detection of nitrite–nitrogen concentration in surface water with ultraviolet–visible spectroscopy.
- Authors
Li, Q.; Liu, R.; Shang, Y.; Wei, Y.; Cui, H.
- Abstract
With the rapid development of the economy, surface water pollution is still serious. Nitrite nitrogen can reflect the degree of surface water pollution. Ultraviolet–visible spectroscopy has the advantages of environmental friendliness, easy operation and real-time online in-situ detection, and has a good application prospect in the detection of nitrite nitrogen content in surface water. In this study, stacking learning and information entropy weighting are introduced on the basis of traditional extreme learning machine, and a stacked extreme learning machine model based on information entropy weight is proposed, aiming at the problem of nitrite nitrogen detection in surface water. In the experiment, 36 nitrite nitrogen aqueous solution samples were prepared, 25 samples were divided into training sets, and the remaining 11 were used as test sets for grouping experiments. In order to further verify the effectiveness of the model under different instrument conditions, the spectra of samples with different concentrations were collected the next day to form a second test set. The experimental results show that the stacked extreme learning machine model based on information entropy weight is superior to the traditional models in measuring accuracy, and can quickly and accurately detect nitrite nitrogen in surface water, providing an effective method for online detection of total organic carbon in surface water quality.
- Subjects
MACHINE learning; ULTRAVIOLET-visible spectroscopy; NITROGEN in water; WATER pollution; ENTROPY (Information theory)
- Publication
International Journal of Environmental Science & Technology (IJEST), 2024, Vol 21, Issue 9, p6653
- ISSN
1735-1472
- Publication type
Article
- DOI
10.1007/s13762-023-05426-2