Journal article
Strategies for efficient machine learning of surrogate drag models from three-dimensional mesoscale computations of shocked particulate flows
International journal of multiphase flow, Vol.108, pp.51-68
11/2018
DOI: 10.1016/j.ijmultiphaseflow.2018.06.013
Abstract
•Cost-effective strategies for developing surrogate models for drag on particles in shocked flows are explored.•A machine learning technique is used to develop the surrogate models from simulations based training data.•The training data are obtained from 3D simulations of shock-particle interaction.•The current study shows that the cost of developing simulation-based surrogate models can be reduced by using the multi-fidelity techniques.
Macroscale simulations of shocked particulate flows rely on closure laws to model momentum transfer between the fluid and dispersed particles phase. Developing closure models from experimental data is expensive. Robust and accurate closures laws can be obtained through surrogate modeling using high-resolution mesoscale simulations. However, development of surrogate models for drag from 3D high-fidelity simulations of shock interaction with clusters of particles can be computationally prohibitive. This paper explores various strategies to efficiently construct surrogate models for drag on particles in the shocked flow. The cost of generating training data is reduced by selecting optimal grid resolutions, particle arrangements in clusters, and size of particle clusters, i.e., by selecting suitable representative volumes (RVEs). Different surrogate modeling strategies such as multi-fidelity and parameter-by-parameter construction approaches are examined. The surrogate models obtained from the different methods are compared to determine the most cost-effective machine learning based surrogate modeling method in the context of shock-particle interactions.
Details
- Title: Subtitle
- Strategies for efficient machine learning of surrogate drag models from three-dimensional mesoscale computations of shocked particulate flows
- Creators
- Pratik Das - Department of Mechanical and Industrial Engineering, The University of Iowa, Iowa City, IA 52242, USAOishik Sen - Department of Mechanical and Industrial Engineering, The University of Iowa, Iowa City, IA 52242, USAK.K Choi - Department of Mechanical and Industrial Engineering, The University of Iowa, Iowa City, IA 52242, USAGustaaf Jacobs - Aerospace Engineering, San Diego State University, San Diego, CA 92115, USAH.S Udaykumar - Department of Mechanical and Industrial Engineering, The University of Iowa, Iowa City, IA 52242, USA
- Resource Type
- Journal article
- Publication Details
- International journal of multiphase flow, Vol.108, pp.51-68
- DOI
- 10.1016/j.ijmultiphaseflow.2018.06.013
- ISSN
- 0301-9322
- eISSN
- 1879-3533
- Publisher
- Elsevier Ltd
- Grant note
- DOI: 10.13039/100000181, name: Air Force Office of Scientific Research, award: FA9550-15-1-0332, SA0000506
- Language
- English
- Date published
- 11/2018
- Academic Unit
- IIHR--Hydroscience and Engineering; Injury Prevention Research Center; Mechanical Engineering
- Record Identifier
- 9984121961602771
Metrics
12 Record Views