TY - JOUR

T1 - Reconstruction of Process Forces in a Five-Axis Milling Center with a LSTM Neural Network in Comparison to a Model-Based Approach

AU - Denkena, Berend

AU - Bergmann, Benjamin

AU - Stoppel, Dennis

N1 - Funding information: Acknowledgments: We thank the DFG for funding this project and our project partner DECKEL MAHO Seebach GmbH. Funding: This research was funded by the German Research Foundation (DFG), grant number DE447/160-1.

PY - 2020/9

Y1 - 2020/9

N2 - Based on the drive signals of a milling center, process forces can be reconstructed. Therefore, a novel approach is presented to reconstruct the process forces with a long short-term memory neural network (LSTM) using drive signals as an input. The LSTM is evaluated and compared to a model-based approach. The latter compensates nonlinearities and disturbances such as friction and inertia. For training of the LSTM, multiple milling processes are considered to enhance the generalizability. Training data is generated by recording drive signals and process forces measured by a dynamometer. The LSTM is then evaluated using a test set, which comprises new process parameters. It is shown that the LSTM has a lower root mean square error (RMSE) in comparison to the model-based approach. Especially, when changing the feed motion direction during milling, the neural network clearly outperforms the model-based approach. Nevertheless, there are processes, where the LSTM induced oscillations, which do not correspond to the measured forces.

AB - Based on the drive signals of a milling center, process forces can be reconstructed. Therefore, a novel approach is presented to reconstruct the process forces with a long short-term memory neural network (LSTM) using drive signals as an input. The LSTM is evaluated and compared to a model-based approach. The latter compensates nonlinearities and disturbances such as friction and inertia. For training of the LSTM, multiple milling processes are considered to enhance the generalizability. Training data is generated by recording drive signals and process forces measured by a dynamometer. The LSTM is then evaluated using a test set, which comprises new process parameters. It is shown that the LSTM has a lower root mean square error (RMSE) in comparison to the model-based approach. Especially, when changing the feed motion direction during milling, the neural network clearly outperforms the model-based approach. Nevertheless, there are processes, where the LSTM induced oscillations, which do not correspond to the measured forces.

KW - Artificial neural network

KW - Machine tools

KW - Process monitoring

UR - http://www.scopus.com/inward/record.url?scp=85092327462&partnerID=8YFLogxK

U2 - 10.3390/jmmp4030062

DO - 10.3390/jmmp4030062

M3 - Article

AN - SCOPUS:85092327462

VL - 4

JO - Journal of Manufacturing and Materials Processing

JF - Journal of Manufacturing and Materials Processing

IS - 3

M1 - 4030062

ER -