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

Publikation: Beiträge in ZeitschriftenZeitschriftenaufsätzeForschungbegutachtet

Authors

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.

OriginalspracheEnglisch
Aufsatznummer1883
ZeitschriftMaterials
Jahrgang14
Ausgabenummer8
Anzahl der Seiten19
ISSN1996-1944
DOIs
PublikationsstatusErschienen - 10.04.2021

Dokumente

DOI

Zuletzt angesehen

Publikationen

  1. How difficult is the adaptation of POS taggers?
  2. Intellectual Property Issues in Space Contracts
  3. Learning how to request using textbooks
  4. Spaces with a temper
  5. Do connectives improve the level of understandability in mathematical reality-based tasks?
  6. Temperature control in Peltier cells comparing sliding mode control and PID controllers
  7. Interplays between relational and instrumental values
  8. Comparison of modeling approaches based on the microstructure of thermally sprayed coatings
  9. rSOESGOPE Method Applied to Four-Tank System Modeling
  10. Non-invariance? An Overstated Problem With Misconceived Causes
  11. Analytics and Intuition in the Process of Selecting Talent
  12. Using rating scales for the assessment of physical self-concept
  13. Principled Interpolation in Normalizing Flows
  14. Learning with summaries
  15. Facing Up to Third Party Liability for Space Activities
  16. Extraterritorial Asylum Processing. The Libya-Niger Emergency Transit Mechanism
  17. Comparison of Backpropagation and Kalman Filter-based Training for Neural Networks
  18. How much can we learn about voluntary climate action from behavior in public goods games?
  19. A dynamic perspective on affect and creativity
  20. Investigating Factors on R estorative Sleep Quality and its Relationship with Personal Work Performance - An Analysis of Diary Data
  21. An introduction to sliding mode control for interdisciplinary education
  22. Leaf trait variation within individuals mediates the relationship between tree species richness and productivity
  23. Sustainability-oriented technology exploration: managerial values, ambidextrous design, and separation drift
  24. Norms and variation in L2 pragmatics
  25. Determinants in the online distribution of digital content
  26. Modeling of 3D fluid-structure-interaction during in-situ hybridization of double-curved fiber-metal-laminates
  27. Co-production of nature's contributions to people
  28. Out of the box
  29. Logistische Lageranalyse und Methodenvalidierung
  30. An Unusual Encounter with Oneself
  31. Introduction to the Special Issue Section
  32. An Approach for Ex-Post-Facto Analysis of Knowledge Graph-Driven Chatbots – The DBpedia Chatbot