Dreaming neural networks for adaptive polishing

Publikation: Beitrag in Buch/Bericht/Sammelwerk/KonferenzbandAufsatz in KonferenzbandForschung

Autoren

  • Marc André Dittrich
  • Bodo Rosenhahn
  • Marcus Magnor
  • Berend Denkena
  • Talash Malek
  • Marco Munderloh
  • Marc Kassubeck
Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Titel des SammelwerksProceedings of the 20th International Conference of the European Society for Precision Engineering and Nanotechnology
UntertitelEUSPEN 2020
Seiten263-266
Seitenumfang4
ISBN (elektronisch)9780995775176
PublikationsstatusVeröffentlicht - 2020
Veranstaltung20th International Conference of the European Society for Precision Engineering and Nanotechnology, EUSPEN 2020 - Geneva, Virtual, Österreich
Dauer: 8 Juni 202012 Juni 2020

Abstract

Mechanical polishing is an important step within the process chain of manufacturing workpieces with high requirements regarding the surface quality, e.g. for optical components. The resulting surface quality depends on several parameters, e.g. the process parameters, the workpiece material, the initial surface roughness and the tool condition. Thus, finding process parameters that lead to the desired surface roughness can be regarded as a complex optimization problem. For this purpose, an artificial neural network (ANN) has been designed and trained with data from polishing experiments. Using a dreaming network approach, the ANN has been enabled to suggest appropriate process parameters under consideration of the initial roughness of the workpiece and the tool condition. The validation experiments showed that the process parameters suggested by the neural network led in 72% to the target roughness within a standard deviation.

ASJC Scopus Sachgebiete

Zitieren

Dreaming neural networks for adaptive polishing. / Dittrich, Marc André; Rosenhahn, Bodo; Magnor, Marcus et al.
Proceedings of the 20th International Conference of the European Society for Precision Engineering and Nanotechnology: EUSPEN 2020. 2020. S. 263-266.

Publikation: Beitrag in Buch/Bericht/Sammelwerk/KonferenzbandAufsatz in KonferenzbandForschung

Dittrich, MA, Rosenhahn, B, Magnor, M, Denkena, B, Malek, T, Munderloh, M & Kassubeck, M 2020, Dreaming neural networks for adaptive polishing. in Proceedings of the 20th International Conference of the European Society for Precision Engineering and Nanotechnology: EUSPEN 2020. S. 263-266, 20th International Conference of the European Society for Precision Engineering and Nanotechnology, EUSPEN 2020, Geneva, Virtual, Österreich, 8 Juni 2020. <https://www.euspen.eu/knowledge-base/ICE20379.pdf>
Dittrich, M. A., Rosenhahn, B., Magnor, M., Denkena, B., Malek, T., Munderloh, M., & Kassubeck, M. (2020). Dreaming neural networks for adaptive polishing. In Proceedings of the 20th International Conference of the European Society for Precision Engineering and Nanotechnology: EUSPEN 2020 (S. 263-266) https://www.euspen.eu/knowledge-base/ICE20379.pdf
Dittrich MA, Rosenhahn B, Magnor M, Denkena B, Malek T, Munderloh M et al. Dreaming neural networks for adaptive polishing. in Proceedings of the 20th International Conference of the European Society for Precision Engineering and Nanotechnology: EUSPEN 2020. 2020. S. 263-266
Dittrich, Marc André ; Rosenhahn, Bodo ; Magnor, Marcus et al. / Dreaming neural networks for adaptive polishing. Proceedings of the 20th International Conference of the European Society for Precision Engineering and Nanotechnology: EUSPEN 2020. 2020. S. 263-266
Download
@inproceedings{50a362fac9384556a19c38c34da0f8f1,
title = "Dreaming neural networks for adaptive polishing",
abstract = "Mechanical polishing is an important step within the process chain of manufacturing workpieces with high requirements regarding the surface quality, e.g. for optical components. The resulting surface quality depends on several parameters, e.g. the process parameters, the workpiece material, the initial surface roughness and the tool condition. Thus, finding process parameters that lead to the desired surface roughness can be regarded as a complex optimization problem. For this purpose, an artificial neural network (ANN) has been designed and trained with data from polishing experiments. Using a dreaming network approach, the ANN has been enabled to suggest appropriate process parameters under consideration of the initial roughness of the workpiece and the tool condition. The validation experiments showed that the process parameters suggested by the neural network led in 72% to the target roughness within a standard deviation.",
keywords = "Control loop, Deep learning, Mechanical polishing, Neural network, Process planning",
author = "Dittrich, {Marc Andr{\'e}} and Bodo Rosenhahn and Marcus Magnor and Berend Denkena and Talash Malek and Marco Munderloh and Marc Kassubeck",
note = "Funding information: [ This research has been funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany's Excellence Strategy within the Cluster of Excellence PhoenixD (EXC 2122, Project ID 390833453).; 20th International Conference of the European Society for Precision Engineering and Nanotechnology, EUSPEN 2020 ; Conference date: 08-06-2020 Through 12-06-2020",
year = "2020",
language = "English",
pages = "263--266",
booktitle = "Proceedings of the 20th International Conference of the European Society for Precision Engineering and Nanotechnology",

}

Download

TY - GEN

T1 - Dreaming neural networks for adaptive polishing

AU - Dittrich, Marc André

AU - Rosenhahn, Bodo

AU - Magnor, Marcus

AU - Denkena, Berend

AU - Malek, Talash

AU - Munderloh, Marco

AU - Kassubeck, Marc

N1 - Funding information: [ This research has been funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany's Excellence Strategy within the Cluster of Excellence PhoenixD (EXC 2122, Project ID 390833453).

PY - 2020

Y1 - 2020

N2 - Mechanical polishing is an important step within the process chain of manufacturing workpieces with high requirements regarding the surface quality, e.g. for optical components. The resulting surface quality depends on several parameters, e.g. the process parameters, the workpiece material, the initial surface roughness and the tool condition. Thus, finding process parameters that lead to the desired surface roughness can be regarded as a complex optimization problem. For this purpose, an artificial neural network (ANN) has been designed and trained with data from polishing experiments. Using a dreaming network approach, the ANN has been enabled to suggest appropriate process parameters under consideration of the initial roughness of the workpiece and the tool condition. The validation experiments showed that the process parameters suggested by the neural network led in 72% to the target roughness within a standard deviation.

AB - Mechanical polishing is an important step within the process chain of manufacturing workpieces with high requirements regarding the surface quality, e.g. for optical components. The resulting surface quality depends on several parameters, e.g. the process parameters, the workpiece material, the initial surface roughness and the tool condition. Thus, finding process parameters that lead to the desired surface roughness can be regarded as a complex optimization problem. For this purpose, an artificial neural network (ANN) has been designed and trained with data from polishing experiments. Using a dreaming network approach, the ANN has been enabled to suggest appropriate process parameters under consideration of the initial roughness of the workpiece and the tool condition. The validation experiments showed that the process parameters suggested by the neural network led in 72% to the target roughness within a standard deviation.

KW - Control loop

KW - Deep learning

KW - Mechanical polishing

KW - Neural network

KW - Process planning

UR - http://www.scopus.com/inward/record.url?scp=85091553318&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85091553318

SP - 263

EP - 266

BT - Proceedings of the 20th International Conference of the European Society for Precision Engineering and Nanotechnology

T2 - 20th International Conference of the European Society for Precision Engineering and Nanotechnology, EUSPEN 2020

Y2 - 8 June 2020 through 12 June 2020

ER -

Von denselben Autoren