Oyebisi, S.O and Owamah, H.I. and Omeje, Maxwell (2023) Application of machine learning algorithm in the internal and external hazards from industrial byproducts. Cleaner Engineering and Technology, 13.
![[img]](http://eprints.covenantuniversity.edu.ng/style/images/fileicons/application_pdf.png) |
PDF
Download (2MB) |
Abstract
Natural radioactive substances that are produced because of industrial processes pose a risk to both the environment and people. An extensive analysis of the radiological properties of industrial byproducts was undertaken in this work, and the risks for both indoor and outdoor environments were assessed based on activity concentrations. The machine learning technique of artificial neural networks was used with various training algorithms to predict the internal and external hazards from these industrial byproducts. The findings demonstrated that, with the exception of incinerated sewage sludge ash, metakaolin, marble powder, nickel slag, pyrite ash, silica fume, steel slag, and glass waste powder, every industrial byproduct examined poses potential indoor and outdoor dangers. All backpropagation training algorithms that were used showed high prediction, according to the neural networks. However, when compared to the Bayesian Regularization and Scaled Conjugate Gradient backpropagation training algorithms, the Levenberg-Marquardt backpropagation technique had the best performance indicators for training, validation, and testing. The results can provide reference information for developing a framework for monitoring hazards and their accompanying precise management.
| Item Type: | Article |
|---|---|
| Uncontrolled Keywords: | Cleaner productionMachine learning algorithm Recycling Responsible consumption and productionWaste valorization |
| Subjects: | T Technology > T Technology (General) T Technology > TA Engineering (General). Civil engineering (General) |
| Divisions: | Faculty of Engineering, Science and Mathematics > School of Civil Engineering and the Environment |
| Depositing User: | AKINWUMI |
| Date Deposited: | 27 Oct 2023 07:42 |
| Last Modified: | 27 Oct 2023 07:42 |
| URI: | http://eprints.covenantuniversity.edu.ng/id/eprint/17449 |
Actions (login required)
 |
View Item |
