Hybrid modelling by machine learning corrections of analytical model predictions towards high-fidelity simulation solutions

Research output: Journal contributionsJournal articlesResearchpeer-review

Standard

Hybrid modelling by machine learning corrections of analytical model predictions towards high-fidelity simulation solutions. / Bock, Frederic E.; Keller, Sören; Huber, Norbert et al.
In: Materials, Vol. 14, No. 8, 1883, 10.04.2021.

Research output: Journal contributionsJournal articlesResearchpeer-review

Harvard

APA

Vancouver

Bibtex

@article{02e117368c55439ea1be8f12b5f2129e,
title = "Hybrid modelling by machine learning corrections of analytical model predictions towards high-fidelity simulation solutions",
abstract = "Within the fields of materials mechanics, the consideration of physical laws in machine learning predictions besides the use of data can enable low prediction errors and robustness as opposed to predictions only based on data. On the one hand, exclusive utilization of fundamental physical relationships might show significant deviations in their predictions compared to reality, due to simplifications and assumptions. On the other hand, using only data and neglecting wellestablished physical laws can create the need for unreasonably large data sets that are required to exhibit low bias and are usually expensive to collect. However, fundamental but simplified physics in combination with a corrective model that compensates for possible deviations, e.g., to experimental data, can lead to physics-based predictions with low prediction errors, also despite scarce data. In this article, it is demonstrated that a hybrid model approach consisting of a physics-based model that is corrected via an artificial neural network represents an efficient prediction tool as opposed to a purely data-driven model. In particular, a semi-analytical model serves as an efficient low-fidelity model with noticeable prediction errors outside its calibration domain. An artificial neural network is used to correct the semi-analytical solution towards a desired reference solution provided by highfidelity finite element simulations, while the efficiency of the semi-analytical model is maintained and the applicability range enhanced. We utilize residual stresses that are induced by laser shock peening as a use-case example. In addition, it is shown that non-unique relationships between model inputs and outputs lead to high prediction errors and the identification of salient input features via dimensionality analysis is highly beneficial to achieve low prediction errors. In a generalization task, predictions are also outside the process parameter space of the training region while remaining in the trained range of corrections. The corrective model predictions show substantially smaller errors than purely data-driven model predictions, which illustrates one of the benefits of the hybrid modelling approach. Ultimately, when the amount of samples in the data set is reduced, the generalization of the physics-related corrective model outperforms the purely data-driven model, which also demonstrates efficient applicability of the proposed hybrid modelling approach to problems where data is scarce.",
keywords = "Analyticalmodel, Artificial neural networks;model correction, Data driven, Feature engineering, Finite elementmodel, Laser shock peening, Machine learning, Physics based, Residual stresses, Engineering",
author = "Bock, {Frederic E.} and S{\"o}ren Keller and Norbert Huber and Benjamin Klusemann",
year = "2021",
month = apr,
day = "10",
doi = "10.3390/ma14081883",
language = "English",
volume = "14",
journal = "Materials",
issn = "1996-1944",
publisher = "MDPI AG",
number = "8",

}

RIS

TY - JOUR

T1 - Hybrid modelling by machine learning corrections of analytical model predictions towards high-fidelity simulation solutions

AU - Bock, Frederic E.

AU - Keller, Sören

AU - Huber, Norbert

AU - Klusemann, Benjamin

PY - 2021/4/10

Y1 - 2021/4/10

N2 - Within the fields of materials mechanics, the consideration of physical laws in machine learning predictions besides the use of data can enable low prediction errors and robustness as opposed to predictions only based on data. On the one hand, exclusive utilization of fundamental physical relationships might show significant deviations in their predictions compared to reality, due to simplifications and assumptions. On the other hand, using only data and neglecting wellestablished physical laws can create the need for unreasonably large data sets that are required to exhibit low bias and are usually expensive to collect. However, fundamental but simplified physics in combination with a corrective model that compensates for possible deviations, e.g., to experimental data, can lead to physics-based predictions with low prediction errors, also despite scarce data. In this article, it is demonstrated that a hybrid model approach consisting of a physics-based model that is corrected via an artificial neural network represents an efficient prediction tool as opposed to a purely data-driven model. In particular, a semi-analytical model serves as an efficient low-fidelity model with noticeable prediction errors outside its calibration domain. An artificial neural network is used to correct the semi-analytical solution towards a desired reference solution provided by highfidelity finite element simulations, while the efficiency of the semi-analytical model is maintained and the applicability range enhanced. We utilize residual stresses that are induced by laser shock peening as a use-case example. In addition, it is shown that non-unique relationships between model inputs and outputs lead to high prediction errors and the identification of salient input features via dimensionality analysis is highly beneficial to achieve low prediction errors. In a generalization task, predictions are also outside the process parameter space of the training region while remaining in the trained range of corrections. The corrective model predictions show substantially smaller errors than purely data-driven model predictions, which illustrates one of the benefits of the hybrid modelling approach. Ultimately, when the amount of samples in the data set is reduced, the generalization of the physics-related corrective model outperforms the purely data-driven model, which also demonstrates efficient applicability of the proposed hybrid modelling approach to problems where data is scarce.

AB - Within the fields of materials mechanics, the consideration of physical laws in machine learning predictions besides the use of data can enable low prediction errors and robustness as opposed to predictions only based on data. On the one hand, exclusive utilization of fundamental physical relationships might show significant deviations in their predictions compared to reality, due to simplifications and assumptions. On the other hand, using only data and neglecting wellestablished physical laws can create the need for unreasonably large data sets that are required to exhibit low bias and are usually expensive to collect. However, fundamental but simplified physics in combination with a corrective model that compensates for possible deviations, e.g., to experimental data, can lead to physics-based predictions with low prediction errors, also despite scarce data. In this article, it is demonstrated that a hybrid model approach consisting of a physics-based model that is corrected via an artificial neural network represents an efficient prediction tool as opposed to a purely data-driven model. In particular, a semi-analytical model serves as an efficient low-fidelity model with noticeable prediction errors outside its calibration domain. An artificial neural network is used to correct the semi-analytical solution towards a desired reference solution provided by highfidelity finite element simulations, while the efficiency of the semi-analytical model is maintained and the applicability range enhanced. We utilize residual stresses that are induced by laser shock peening as a use-case example. In addition, it is shown that non-unique relationships between model inputs and outputs lead to high prediction errors and the identification of salient input features via dimensionality analysis is highly beneficial to achieve low prediction errors. In a generalization task, predictions are also outside the process parameter space of the training region while remaining in the trained range of corrections. The corrective model predictions show substantially smaller errors than purely data-driven model predictions, which illustrates one of the benefits of the hybrid modelling approach. Ultimately, when the amount of samples in the data set is reduced, the generalization of the physics-related corrective model outperforms the purely data-driven model, which also demonstrates efficient applicability of the proposed hybrid modelling approach to problems where data is scarce.

KW - Analyticalmodel

KW - Artificial neural networks;model correction

KW - Data driven

KW - Feature engineering

KW - Finite elementmodel

KW - Laser shock peening

KW - Machine learning

KW - Physics based

KW - Residual stresses

KW - Engineering

UR - http://www.scopus.com/inward/record.url?scp=85104473182&partnerID=8YFLogxK

U2 - 10.3390/ma14081883

DO - 10.3390/ma14081883

M3 - Journal articles

C2 - 33920078

AN - SCOPUS:85104473182

VL - 14

JO - Materials

JF - Materials

SN - 1996-1944

IS - 8

M1 - 1883

ER -

Documents

DOI

Recently viewed

Activities

  1. GIS program – Geographischer Informationssyteme
  2. Performance under Climatic Conditions
  3. 13th IFAC and IEEE Conference on Programmable Devices and Embedded Systems - PDeS 2015
  4. 13th IWH-CIREQ Macroeconometric Workshop: Macro-econometrics and Panel Data - IWH-CIREQ 2012
  5. Projektantrag DFG
  6. The Advance of Diagnosis Chatbots: Should We First Avoid Distrust Before We Focus on Trust?
  7. Es geht um
  8. Conference "FICTION AND SIMULATION: Theories, Methods, Practices" - 2015
  9. Ecology & Evolution, Lecture series 2012
  10. Extending Working Lives in Organizations: The Later Life Workplace Index for Successful Management of an Aging Workforce
  11. PEER Workshop, Aalto University, Helsinki
  12. Offizieller Auftakt des BISS-Programms
  13. Authenticity in modelling test items
  14. Building the 5-Dollar-Neuron. A Genealogy of the ›Biological Computer‹ as Research Program and Metaphor (Konferenz "Neural Initiatives: Neuropolitics, Technoscience, and the Dynamic Brain")
  15. The Imaginary Dimension of Fragile Statehood: Syria and the Struggle for the "Rightful" Polity
  16. Video or Text Cases in Problem-Oriented or Direct Instructional Settings for Preservice Teachers?
  17. Towards a sustainable use of indium – Relating technology options and governance interventions for dealing with indium criticality
  18. Automatic Detection and Classification of State Heads and Common People?
  19. Ja­son W. Moo­re in Con­ver­sa­ti­on with Erich Hörl
  20. Antitrust and Beyond - The Democratic Task of Antitrust Law in the Light of Heinrich Kronstein's Work
  21. CyberKant, a Timely Response to the Eclipse of Reason by Mechanical Rationality - 2020
  22. How is Research Creation as Other Knowledge?

Publications

  1. Theorie des Quantum Computings
  2. Petri net based EMIS-mappers for flexible manufacturing systems
  3. Determination of 10 particle-associated multiclass polar and semi-polar pesticides from small streams using accelerated solvent extraction
  4. Performance concepts and performance theory
  5. Convergence of adaptive learning and expectational stability
  6. Introduction
  7. Analysis of the mechanical properties of an arc-sprayed WC-FeCSiMn coating
  8. Implementing the Kyoto Protocol without Russia
  9. On the role of linguistic features for comprehension and learning from STEM texts. A meta-analysis
  10. Do guided internet-based interventions result in clinically relevant changes for patients with depression?
  11. Optimal dynamic scale and structure of a multi-pollution economy
  12. Individual Scans Fusion in Virtual Knowledge Base for Navigation of Mobile Robotic Group with 3D TVS
  13. How do students and teachers deal with mathematical modelling problems?
  14. In situ synchrotron diffraction of the solidification of Mg4Y3Nd
  15. Hermann Stutte
  16. Different facets of tree sapling diversity influence browsing intensity by deer dependent on spatial scale
  17. An innovative efficiency of incubator to enhance organization supportive business using machine learning approach
  18. Explorations in social spaces
  19. Study of Single Filament Dielectric Barrier Discharge in Argon
  20. It is not what it is
  21. Estimation of minimal data sets sizes for machine learning predictions in digital mental health interventions
  22. Principled Interpolation in Normalizing Flows
  23. Comparison of Backpropagation and Kalman Filter-based Training for Neural Networks
  24. Evidence for singlet state β cleavage in the photoreaction of α-(2,6-dimethoxyphenoxy)-acetophenone inferred from time-resolved CIDNP spectroscopy