Louis, Patrick ; Cibulski, Lena ; Suschnigg, Josef ; Marth, Edmund ; Mitterhofer, Hubert ; Kohlhammer, Jörn ; Schreck, Tobias ; Mutlu, Belgin (2024)
Visual Exploration and Analysis of Simulation and Testing Data in Motor Engineering.
In: IEEE Computer Graphics and Applications
doi: 10.1109/MCG.2024.3392969
Artikel, Bibliographie
Kurzbeschreibung (Abstract)
End-of-line tests and defect detection are vital for ensuring the reliability of electric motors. However, automated defect detection methods, e.g., data-driven approaches, face challenges due to the limited availability of real data from failed motors. Simulated data, though beneficial, lacks the complexity of real motors, impacting the performance of these methods when applied to actual observations. To tackle this challenge, we introduce a visual analysis tool designed to facilitate the analysis of measured and simulated data, presented in the form of time series data. This tool helps identify domain-invariant features and evaluate simulation data accuracy, assisting in selecting training data for reliable automated defect detection in real-world scenarios. The main contribution of this work is a design proposal based on visual design principles, specifically tailored to address the unique requirements of electric motor professionals. The visual design is validated by findings from a think-aloud study with specialized engineers.
| Typ des Eintrags: | Artikel |
|---|---|
| Erschienen: | 2024 |
| Autor(en): | Louis, Patrick ; Cibulski, Lena ; Suschnigg, Josef ; Marth, Edmund ; Mitterhofer, Hubert ; Kohlhammer, Jörn ; Schreck, Tobias ; Mutlu, Belgin |
| Art des Eintrags: | Bibliographie |
| Titel: | Visual Exploration and Analysis of Simulation and Testing Data in Motor Engineering |
| Sprache: | Englisch |
| Publikationsjahr: | 2024 |
| Ort: | Piscataway, NY |
| Verlag: | IEEE |
| Titel der Zeitschrift, Zeitung oder Schriftenreihe: | IEEE Computer Graphics and Applications |
| Kollation: | 13 Seiten |
| DOI: | 10.1109/MCG.2024.3392969 |
| Kurzbeschreibung (Abstract): | End-of-line tests and defect detection are vital for ensuring the reliability of electric motors. However, automated defect detection methods, e.g., data-driven approaches, face challenges due to the limited availability of real data from failed motors. Simulated data, though beneficial, lacks the complexity of real motors, impacting the performance of these methods when applied to actual observations. To tackle this challenge, we introduce a visual analysis tool designed to facilitate the analysis of measured and simulated data, presented in the form of time series data. This tool helps identify domain-invariant features and evaluate simulation data accuracy, assisting in selecting training data for reliable automated defect detection in real-world scenarios. The main contribution of this work is a design proposal based on visual design principles, specifically tailored to address the unique requirements of electric motor professionals. The visual design is validated by findings from a think-aloud study with specialized engineers. |
| Freie Schlagworte: | Exploratory visualization, Defect detection, Time series analysis |
| Fachbereich(e)/-gebiet(e): | 20 Fachbereich Informatik 20 Fachbereich Informatik > Graphisch-Interaktive Systeme |
| Hinterlegungsdatum: | 17 Mai 2024 09:18 |
| Letzte Änderung: | 03 Jun 2024 09:36 |
| PPN: | 518773558 |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum