CaltechAUTHORS
  A Caltech Library Service

A Gaussian moment method and its augmentation via LSTM recurrent neural networks for the statistics of cavitating bubble populations

Bryngelson, Spencer H. and Charalampopoulos, Alexis and Sapsis, Themistoklis P. and Colonius, Tim (2020) A Gaussian moment method and its augmentation via LSTM recurrent neural networks for the statistics of cavitating bubble populations. International Journal of Multiphase Flow, 127 . Art. No. 103262. ISSN 0301-9322. https://resolver.caltech.edu/CaltechAUTHORS:20200309-084519190

[img] PDF - Submitted Version
See Usage Policy.

541Kb

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20200309-084519190

Abstract

Phase-averaged dilute bubbly flow models require high-order statistical moments of the bubble population. The method of classes, which directly evolve bins of bubbles in the probability space, are accurate but computationally expensive. Moment-based methods based upon a Gaussian closure present an opportunity to accelerate this approach, particularly when the bubble size distributions are broad (polydisperse). For linear bubble dynamics a Gaussian closure is exact, but for bubbles undergoing large and nonlinear oscillations, it results in a large error from misrepresented higher-order moments. Long short-term memory recurrent neural networks, trained on Monte Carlo truth data, are proposed to improve these model predictions. The networks are used to correct the low-order moment evolution equations and improve prediction of higher-order moments based upon the low-order ones. Results show that the networks can reduce model errors to less than 1% of their unaugmented values.


Item Type:Article
Related URLs:
URLURL TypeDescription
https://doi.org/10.1016/j.ijmultiphaseflow.2020.103262DOIArticle
https://arxiv.org/abs/1912.04450arXivDiscussion Paper
ORCID:
AuthorORCID
Bryngelson, Spencer H.0000-0003-1750-7265
Colonius, Tim0000-0003-0326-3909
Additional Information:© 2020 Elsevier Ltd. Received 10 December 2019, Revised 21 February 2020, Accepted 2 March 2020, Available online 6 March 2020. The US Office of Naval Research supported this work under MURI grant N0014-17-1-2676.
Funders:
Funding AgencyGrant Number
Office of Naval Research (ONR)N0014-17-1-2676
Subject Keywords:Bubbly flow; phase averaging; moment methods; recurrent neural networks
Record Number:CaltechAUTHORS:20200309-084519190
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20200309-084519190
Official Citation:Spencer H. Bryngelson, Alexis Charalampopoulos, Themistoklis P. Sapsis, Tim Colonius, A Gaussian moment method and its augmentation via LSTM recurrent neural networks for the statistics of cavitating bubble populations, International Journal of Multiphase Flow, Volume 127, 2020, 103262, ISSN 0301-9322, https://doi.org/10.1016/j.ijmultiphaseflow.2020.103262. (http://www.sciencedirect.com/science/article/pii/S0301932219309644)
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:101766
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:09 Mar 2020 16:27
Last Modified:20 Mar 2020 20:51

Repository Staff Only: item control page