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

Publications

  1. Linked Accomplishment Of Order Management And Production Planning And Control. An Integrated Model-based Approach
  2. Conjunctive cohesion in English language EU documents - A corpus-based analysis and its implications
  3. Development and application of a laboratory flux measurement system (LFMS) for the investigation of the kinetics of mercury emissions from soils
  4. Nmap: A novel neighborhood preservation space-filling algorithm
  5. Nest site selection and the effects of land use in a multi-scale approach on the distribution of a passerine in an island arid environment
  6. Phosphorus uptake from struvite is modulated by the nitrogen form applied
  7. Examining how AI capabilities can foster organizational performance in public organizations
  8. Low working memory reduces the use of mental contrasting
  9. Deterministic Pod Repositioning in Robotic Mobile Fulfillment Systems
  10. A blueprint for mapping and modelling ecosystem services
  11. Ontology-Guided, Hybrid Prompt Learning for Generalization in Knowledge Graph Question Answering
  12. Systemnahe Programmierung
  13. Exploring intrinsic, instrumental and relational values for sustainable management of social-ecological systems
  14. Global patterns of ecologically unequal exchange
  15. 9th challenge on question answering over linked data (QALD-9)
  16. Acting in the Name of Others
  17. Learning Analytics and Digital Badges
  18. Take the money and run? Implementation and disclosure of environmentally-oriented crowdfunding projects
  19. Smart Devices Evaluation and Dynamic Cognitive Assistance System for Repair Processes in Production
  20. Future-proofing ecosystem restoration through enhancing adaptive capacity
  21. Second-Order Sliding Mode Control with State and Disturbance Estimation for a Permanent Magnet Linear Motor
  22. Operationalising the leverage points perspective for empirical research
  23. Differentiated Instruction Around the World - A Global Inclusive Insight
  24. Strategy execution in hospitals
  25. To use or not to use learning data
  26. Better performance of organic than conventional tomato varieties in single and mixed cropping
  27. Reconfiguring Desecuritization
  28. The language of situated joint activity: Social virtual reality and language learning in virtual exchange
  29. Introduction: Toward a business administration for the 21st century
  30. Experimental and numerical analysis of refill friction stir spot welding of thin AA7075-T6 sheets
  31. Like! You saved #energy today. Fostering Energy Efficiency in Buildings – The implementation of social media patterns as symbols in Building Management Systems‘ Graphical User Interfaces using Peirce’s semeiosis as a communication concept
  32. Application of novel constrained friction processing method to produce fine grained biomedical Mg-Zn-Ca alloy
  33. Allometric equations for maximum filtration rate in blue mussels Mytilus edulis and importance of condition index
  34. Export scope and characteristics of destination countries
  35. The effect of storage medium on the elution of monomers from composite materials
  36. "Wen feiern wir denn eigentlich?"
  37. Dispersal and diversity–unifying scale-dependent relationships within the neutral theory
  38. Carbon performance and disclosure
  39. Effect of salinity on growth of mussels, Mytilus edulis, with special reference to Great Belt (Denmark)