Szczegóły publikacji
Opis bibliograficzny
Optimizing remaining useful life prediction: a feature engineering approach / Paweł KNAP, Urszula Jachymczyk, Krzysztof LALIK // W: ICCC 2024 [Dokument elektroniczny] : 25th International Carpathian Control Conference : 22–24 May 2024, Krynica-Zdrój, Poland : proceedings / ed. Andrzej Kot. — Wersja do Windows. — Dane tekstowe. — Piscataway : IEEE, cop. 2024. — Dod. ISBN: 979-8-3503-5069-2, 979-8-3503-5071-5. — e-ISBN: 979-8-3503-5070-8. — S. [1–5]. — Wymagania systemowe: Adobe Reader. — Bibliogr. s. [4–5], Abstr. — Publikacja dostępna online od: 2024-07-01
Autorzy (3)
Słowa kluczowe
Dane bibliometryczne
| ID BaDAP | 155013 |
|---|---|
| Data dodania do BaDAP | 2024-09-11 |
| Tekst źródłowy | URL |
| DOI | 10.1109/ICCC62069.2024.10569235 |
| Rok publikacji | 2024 |
| Typ publikacji | materiały konferencyjne (aut.) |
| Otwarty dostęp |  |
| Wydawca | Institute of Electrical and Electronics Engineers (IEEE) |
Abstract
Machine learning techniques have recently brought many improvements in the field of machine’s Prognostics and Health Management (PHM). The main objectives include identification of degradation patterns indicating potential issues, forecasting Remaining Useful Life (RUL), optimizing undertaking maintenance actions, and minimizing downtime. In this paper, we focused on creating a successful RUL model, which allows predicting the amount of time when a machine, or its element, will function effectively before failure. The success of such an intelligent health assessment model depends not only on a chosen algorithm but also thorough data preprocessing and feature selection. Dimensionality reduction algorithms can help prevent the so-called’curse of dimensionality,’ which refers to the challenges when working with high-dimensional data. A large number of features create the risk of being very sparse, instances are far away from each other, which affects the reliability of predictions. We have chosen one of the most popular dimensionality reduction techniques - Principal Component Analysis (PCA), which projects data down to a determined number of dimensions. Gathering data for RUL is quite challenging because it requires measurements while running the machine to failure; therefore, it has been decided to use bearing open-source dataset prepared in an environment well-adjusted for this demanding task. ©2024 IEEE.
