Increased Reliability of Draw-In Prediction in a Single Stage Deep-Drawing Operation via Transfer Learning

Lea Wollschlaeger; Christine Heinzel; Sebastian Thiery; Mazhar Zein El Abdine; Noomane Ben Khalifa; Jens Heger

doi:10.1016/j.procir.2024.10.086

Increased Reliability of Draw-In Prediction in a Single Stage Deep-Drawing Operation via Transfer Learning

Research output: Journal contributions › Conference article in journal › Research › peer-review

Standard

Increased Reliability of Draw-In Prediction in a Single Stage Deep-Drawing Operation via Transfer Learning. / Wollschlaeger, Lea ; Heinzel, Christine ; Thiery, Sebastian et al.
In: Procedia CIRP, Vol. 130, 01.01.2024, p. 270-275.

Research output: Journal contributions › Conference article in journal › Research › peer-review

Bibtex

@article{116e0395f38c4d9ab4d3aee441a76bd7,

title = "Increased Reliability of Draw-In Prediction in a Single Stage Deep-Drawing Operation via Transfer Learning",

abstract = "In sheet metal forming operations, finite element simulations and experimental works are used to evaluate predictions on different parameter settings. During the manufacturing process, there often exists a discrepancy of expected outcomes due to varying material properties. With the aim to save simulation and experimental resources, this paper provides a reliable transfer learning model suiting the deep-drawing case, where a model is pre-trained on simulation data, neurons are frozen in different layers and is then fine-tuned on real data. This model is evaluated in its behavior by gradually learning on different number of real data points as well as simulation data points.",

keywords = "Artificial Intelligence, Machine Learning, Artificial neural network, Transfer Learning, Industrial applications, Deep-drawing, Engineering",

author = "Lea Wollschlaeger and Christine Heinzel and Sebastian Thiery and Abdine, {Mazhar Zein El} and Khalifa, {Noomane Ben} and Jens Heger",

note = "Publisher Copyright: {\textcopyright} 2024 The Authors.; 57th CIRP Conference on Manufacturing Systems - CIRP CMS 2024 : Speeding up manufacturing, CIRP CMS '24 ; Conference date: 29-05-2024 Through 31-05-2024",

year = "2024",

month = jan,

day = "1",

doi = "10.1016/j.procir.2024.10.086",

language = "English",

volume = "130",

pages = "270--275",

journal = "Procedia CIRP",

issn = "2212-8271",

publisher = "Elsevier B.V.",

url = "https://www.cirpcms2024.org/",

}

RIS

TY - JOUR

T1 - Increased Reliability of Draw-In Prediction in a Single Stage Deep-Drawing Operation via Transfer Learning

AU - Wollschlaeger, Lea

AU - Heinzel, Christine

AU - Thiery, Sebastian

AU - Abdine, Mazhar Zein El

AU - Khalifa, Noomane Ben

AU - Heger, Jens

N1 - Conference code: 57

PY - 2024/1/1

Y1 - 2024/1/1

N2 - In sheet metal forming operations, finite element simulations and experimental works are used to evaluate predictions on different parameter settings. During the manufacturing process, there often exists a discrepancy of expected outcomes due to varying material properties. With the aim to save simulation and experimental resources, this paper provides a reliable transfer learning model suiting the deep-drawing case, where a model is pre-trained on simulation data, neurons are frozen in different layers and is then fine-tuned on real data. This model is evaluated in its behavior by gradually learning on different number of real data points as well as simulation data points.

AB - In sheet metal forming operations, finite element simulations and experimental works are used to evaluate predictions on different parameter settings. During the manufacturing process, there often exists a discrepancy of expected outcomes due to varying material properties. With the aim to save simulation and experimental resources, this paper provides a reliable transfer learning model suiting the deep-drawing case, where a model is pre-trained on simulation data, neurons are frozen in different layers and is then fine-tuned on real data. This model is evaluated in its behavior by gradually learning on different number of real data points as well as simulation data points.

KW - Artificial Intelligence

KW - Machine Learning

KW - Artificial neural network

KW - Transfer Learning

KW - Industrial applications

KW - Deep-drawing

KW - Engineering

UR - http://www.scopus.com/inward/record.url?scp=85213035295&partnerID=8YFLogxK

U2 - 10.1016/j.procir.2024.10.086

DO - 10.1016/j.procir.2024.10.086

M3 - Conference article in journal

VL - 130

SP - 270

EP - 275

JO - Procedia CIRP

JF - Procedia CIRP

SN - 2212-8271

T2 - 57th CIRP Conference on Manufacturing Systems - CIRP CMS 2024

Y2 - 29 May 2024 through 31 May 2024

ER -

Data based analysis of order processing strategies to support the positioning between conflicting economic and logistic objectives

Maier, J. T., Heuer, T., Stoffersen, H., Nyhuis, P. & Schmidt, M., 01.01.2022, In: Procedia CIRP. 107, p. 332-337 6 p.

Research output: Journal contributions › Conference article in journal › Research › peer-review

Effect of Thermal Expansion on the Dynamics of Rolling-element Bearing

Ambrozkiewicz, B., Gassner, A., Meier, N., Litak, G. & Georgiadis, A., 22.09.2022, In: Procedia CIRP. 112, p. 151-155 5 p.

Research output: Journal contributions › Conference article in journal › Research › peer-review