Tool Wear Prediction Using Smart Data and Advanced Change Point Detection Techniques for Optimized Replacement Timing in Manufacturing Systems
Artykuł w czasopiśmie
MNiSW
20
Lista 2024
| Status: | |
| Autorzy: | Janik Mateusz, Krzempek Konrad, Sobecki Piotr, Mazurkiewicz Dariusz, Żabiński Tomasz, Piecuch Grzegorz |
| Dyscypliny: | |
| Aby zobaczyć szczegóły należy się zalogować. | |
| Rok wydania: | 2025 |
| Wersja dokumentu: | Drukowana | Elektroniczna |
| Język: | angielski |
| Numer czasopisma: | 24 |
| Wolumen/Tom: | 59 |
| Strony: | 137 - 142 |
| Efekt badań statutowych | NIE |
| Finansowanie: | This work was supported by the individual research grant No. FD-20/IM-5/072/2024, awarded for the scientific discipline of Mechanical Engineering, Lublin University of Technology, Poland. |
| Materiał konferencyjny: | NIE |
| Publikacja OA: | TAK |
| Licencja: | |
| Sposób udostępnienia: | Witryna wydawcy |
| Wersja tekstu: | Ostateczna wersja opublikowana |
| Czas opublikowania: | W momencie opublikowania |
| Data opublikowania w OA: | 9 grudnia 2025 |
| Abstrakty: | angielski |
| This publication addresses a key challenge in predictive maintenance by proposing a novel approach to tool wear prediction based on change-point detection in signal characteristics. Despite extensive studies on tool wear monitoring, many existing methods lack accuracy in identifying early wear stages under real-time manufacturing conditions. To bridge this gap, this study employs the Pruned Exact Linear Time (PELT) algorithm combined with bandpass filtering and wavelet-based signal energy analysis. The proposed method enables precise detection of tool wear progression by identifying characteristic frequency shifts and abrupt signal changes. Experimental results demonstrate the effectiveness of this approach in forecasting optimal tool replacement timing, reducing maintenance costs, and enhancing manufacturing system reliability. This contribution offers a practical and scalable solution, with potential for integration into real-time machine health monitoring frameworks. |
