Szczegóły publikacji
Opis bibliograficzny
Hybrid approach of neural networks and analog-based methods for industrial assessment of technogenic deposits / Artem Pavlychenko, Dagmara LEWICKA, Ivan Miroshnykov, Serhii Dybrin, Andrii Pererva, Roman DYCHKOVSKYI // W: Proceedings of the DIM-ESEE Conference [Dokument elektroniczny] : Dubrovnik, October 15th – 17th, 2025 / eds. Dalibor Kuhinek, [et al.]. — Wersja do Windows. — Dane tekstowe. — Zagreb : University of Zagreb. Faculty of Mining, Geology and Petroleum Engineering, 2025. — ( Proceedings of the DIM-ESEE Conference ; ISSN 3102-3517 ; vol. 1 ). — S. MR13–MR20. — Wymagania systemowe: Adobe Reader. — Bibliogr. s. MR19, Abstr. — R. Dychkovskyi - dod. afiliacja: Dnipro University of Technology, Dnipro, Ukraine
Autorzy (6)
- Pavlychenko Artem
- AGHLewicka Dagmara
- Miroshnykov Ivan
- Dybrin Serhii
- Pererva Andrii
- AGHDychkovskyi Roman
Słowa kluczowe
Dane bibliometryczne
| ID BaDAP | 166745 |
|---|---|
| Data dodania do BaDAP | 2026-03-31 |
| Tekst źródłowy | URL |
| Rok publikacji | 2025 |
| Typ publikacji | materiały konferencyjne (aut.) |
| Otwarty dostęp |  |
| Creative Commons | |
| Czasopismo/seria | Proceedings of the DIM-ESEE Conference |
Abstract
Technogenic deposits, originating from mining and industrial activities, are increasingly recognized as valuable secondary sources of critical raw materials. Traditional analog-based assessment methods offer practical tools for evaluating their industrial potential but often struggle with the complexity and heterogeneity of such formations. This study proposes a hybrid methodology that integrates analog-based approaches with artificial neural networks (ANNs) to enhance the accuracy and reliability of technogenic deposit assessments. Analog methods are first employed to establish baseline parameters from historical data, which then serve as training inputs for a multilayer perceptron (MLP) neural network. The model demonstrated strong predictive performance, achieving classification accuracies above 90% on validation datasets. Sensitivity analysis identified pollutant concentration, mineral composition, and particle size distribution as the most influential factors in determining industrial suitability. By combining empirical knowledge with machine learning, the hybrid approach enables robust, adaptive, and scalable evaluations, even under conditions of incomplete or uncertain data. The results confirm its potential to improve decision-making in resource recovery, environmental remediation, and sustainable development. This framework contributes to the digital transformation of resource management and supports the responsible exploitation of technogenic deposits within circular economy principles.
