TY - JOUR
T1 - Application of Deep Neural Networks for the Parameter Identifications of Lumped and Distributed Parameter Models Under Severe Noises and Various Initial Values
AU - Song, Byoung Gyu
AU - Kang, Namcheol
N1 - Publisher Copyright:
© 2023, Krishtel eMaging Solutions Private Limited.
PY - 2023
Y1 - 2023
N2 - Purpose: Identifying unknown parameters has long been significant in the field of engineering, as they can be used for fault diagnosis and the development of numerical models. This study aims to determine the unknown parameters of mechanical models by applying deep neural networks (DNNs) using frequency–response functions (FRFs). Methods: The proposed network consists of several DNNs that estimate unknown parameters. The inputs of the DNNs requires the initial values of the unknowns. The estimated parameters by the DNNs were used to calculate the outputs. The magnitude and phase of the FRF obtained from an experiment were used as target data. To optimize the performance of the DNNs, Bayesian optimization is employed to search for appropriate weight factors in the loss function and hyperparameters in the DNNs. Additionally, the estimated parameters were adopted as the optimal parameters after the difference between the output and target data satisfied the stopping criterion. Results: The performance of the DNN-based method was confirmed for lumped and distributed parameter models with noise, which could significantly challenge the parameter identification. The proposed method showed high accuracy within 5% for both models and was validated in terms of convergence by simulations and experiments. Furthermore, we compared the proposed method with the conventional optimization methods across various noise levels and initial values. Conclusion: The proposed method successfully estimated unknown parameters under severe noise and initial conditions compared with the other methods.
AB - Purpose: Identifying unknown parameters has long been significant in the field of engineering, as they can be used for fault diagnosis and the development of numerical models. This study aims to determine the unknown parameters of mechanical models by applying deep neural networks (DNNs) using frequency–response functions (FRFs). Methods: The proposed network consists of several DNNs that estimate unknown parameters. The inputs of the DNNs requires the initial values of the unknowns. The estimated parameters by the DNNs were used to calculate the outputs. The magnitude and phase of the FRF obtained from an experiment were used as target data. To optimize the performance of the DNNs, Bayesian optimization is employed to search for appropriate weight factors in the loss function and hyperparameters in the DNNs. Additionally, the estimated parameters were adopted as the optimal parameters after the difference between the output and target data satisfied the stopping criterion. Results: The performance of the DNN-based method was confirmed for lumped and distributed parameter models with noise, which could significantly challenge the parameter identification. The proposed method showed high accuracy within 5% for both models and was validated in terms of convergence by simulations and experiments. Furthermore, we compared the proposed method with the conventional optimization methods across various noise levels and initial values. Conclusion: The proposed method successfully estimated unknown parameters under severe noise and initial conditions compared with the other methods.
KW - Deep neural network (DNN)
KW - Frequency response function (FRF)
KW - Impact hammer test
KW - Lumped and distributed parameter model
KW - Parameter identification
UR - http://www.scopus.com/inward/record.url?scp=85165613967&partnerID=8YFLogxK
U2 - 10.1007/s42417-023-01074-5
DO - 10.1007/s42417-023-01074-5
M3 - Article
AN - SCOPUS:85165613967
SN - 2523-3920
JO - Journal of Vibration Engineering and Technologies
JF - Journal of Vibration Engineering and Technologies
ER -