skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Deep Learning for Presumed Probability Density Function Models

Abstract

In this work, we use machine learning (ML) techniques to develop presumed probability density function (PDF) models for large eddy simulations (LES) of reacting flows. The joint sub-filter PDF of mixture fraction and progress variable is modeled using various ML algorithms and commonly used analytical models. The ML algorithms evaluated in the work are representative of three major classes of ML techniques: traditional ensemble methods (random forests), deep learning (deep neural network (DNN)s), and generative learning (conditional variational autoencoder (CVAE)).

Authors:
 [1];  [1];  [1];  [2];  [1]
  1. National Renewable Energy Laboratory (NREL), Golden, CO (United States)
  2. Lawrence Berkely National Laboratory
Publication Date:
Research Org.:
National Renewable Energy Lab. (NREL), Golden, CO (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Exascale Computing Project (ECP); USDOE National Nuclear Security Administration (NNSA)
OSTI Identifier:
1562859
Report Number(s):
NREL/JA-2C00-73038
DOE Contract Number:  
AC36-08GO28308
Resource Type:
Journal Article
Journal Name:
Combustion and Flame
Additional Journal Information:
Journal Volume: 208
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; large eddy simulation; presumed probability density function; low-swirl burner; machine learning

Citation Formats

Henry de Frahan, Marc, Yellapantula, Shashank, King, Ryan N, Day, Marc S., and Grout, Ray W. Deep Learning for Presumed Probability Density Function Models. United States: N. p., 2019. Web. doi:10.1016/j.combustflame.2019.07.015.
Henry de Frahan, Marc, Yellapantula, Shashank, King, Ryan N, Day, Marc S., & Grout, Ray W. Deep Learning for Presumed Probability Density Function Models. United States. doi:10.1016/j.combustflame.2019.07.015.
Henry de Frahan, Marc, Yellapantula, Shashank, King, Ryan N, Day, Marc S., and Grout, Ray W. Wed . "Deep Learning for Presumed Probability Density Function Models". United States. doi:10.1016/j.combustflame.2019.07.015.
@article{osti_1562859,
title = {Deep Learning for Presumed Probability Density Function Models},
author = {Henry de Frahan, Marc and Yellapantula, Shashank and King, Ryan N and Day, Marc S. and Grout, Ray W},
abstractNote = {In this work, we use machine learning (ML) techniques to develop presumed probability density function (PDF) models for large eddy simulations (LES) of reacting flows. The joint sub-filter PDF of mixture fraction and progress variable is modeled using various ML algorithms and commonly used analytical models. The ML algorithms evaluated in the work are representative of three major classes of ML techniques: traditional ensemble methods (random forests), deep learning (deep neural network (DNN)s), and generative learning (conditional variational autoencoder (CVAE)).},
doi = {10.1016/j.combustflame.2019.07.015},
journal = {Combustion and Flame},
number = ,
volume = 208,
place = {United States},
year = {2019},
month = {8}
}