Szczegóły publikacji
Opis bibliograficzny
AI-based data mining approach to control the environmental impact of conventional energy technologies / Katarzyna SZRAMOWIAT-SALA, Roch Penkala, Jiří Horák, Kamil Krpec, František Hopan, Jiří Ryšavý, Karel Borovec, Jerzy GÓRECKI // Journal of Cleaner Production ; ISSN 0959-6526. — 2024 — vol. 472 art. no. 143473, s. 1–10. — Bibliogr. s. 9–10, Abstr. — Publikacja dostępna online od: 2024-08-28
Autorzy (8)
- AGHSzramowiat-Sala Katarzyna
- AGHPenkala Roch
- Horák Jiří
- Krpec Kamil
- Hopan Frantisek
- Ryšavý Jiří
- Borovec Karel
- AGHGórecki Jerzy
Słowa kluczowe
Dane bibliometryczne
| ID BaDAP | 155277 |
|---|---|
| Data dodania do BaDAP | 2024-10-07 |
| Tekst źródłowy | URL |
| DOI | 10.1016/j.jclepro.2024.143473 |
| Rok publikacji | 2024 |
| Typ publikacji | artykuł w czasopiśmie |
| Otwarty dostęp |  |
| Creative Commons | |
| Czasopismo/seria | Journal of Cleaner Production |
Abstract
Environmental pollution remains one of the foremost existential threats to human well-being, despite the concerted efforts and implementation of various programmes aimed at fostering cleaner air. The contemporary global economic and energy landscape, characterised by multifaceted challenges, has undeniably hindered the efficacy of efforts to kerb air pollutant emissions. Solid fuels persist as primary sources of energy production in numerous countries, serving both the residential and industrial sectors. However, combustion of such fuels, particularly within domestic heating units (DHUs), engenders the release of a diverse array of organic compounds characterised by intricate structures and potent mutagenic and environmentally hazardous properties. However, the combustion process, if properly regulated, can be carried out in an environmentally sustainable manner. The intricate interplay of myriad factors that influence the composition and quality of chimney flue gases underscores the complexity inherent in controlling the combustion process. Artificial intelligence (AI) has emerged as a versatile tool with applications that span various domains, including environmental monitoring systems. In this study, we posit the utilisation of artificial neural networks (ANNs) as a sophisticated data mining technique to control the emission of flue gases contingent on the specific boiler and fuel utilised. Feed forward predictive models with back propagation were utilized for AI-based data mining aiming at the prediction of the concentration of flue gas components. The highest coefficients of model fit goodness were obtained for CO2, NOx and SO2 with R2 equal to 0.99, 0.98 and 0.99, respectively. The study demonstrated the feasibility and effectiveness of using AI-based data mining to predict emissions from conventional energy technologies. By leveraging the predictive capabilities of ANNs, it is possible to significantly reduce the environmental impact of solid fuel combustion, contributing to cleaner air and improved public health. © 2024 The Authors
