Machine learning for combustion system control and simulation

ORAL

Abstract

Real-world combustion systems are highly complex with scales that span many orders of magnitude making them particularly challenging for numerical simulations. Two large challenges are the numerical simulation and control of these systems in engineering applications under realistic conditions. Machine learning has emerged over the past 5+ years to show immense promise in data science applications and also in many real-world engineering applications. Here, we discuss the recent application of various machine learning techniques to assist in the development of control strategies for compression ignition engines using deep reinforcement learning and also the approximation of chemical kinetics mechanisms using supervised learning for applications in computational fluid dynamics codes.

*A portion of this research was conducted as part of the Co-Optimization of Fuels \& Engines (Co-Optima) project sponsored by the U.S. Department of Energy Office of Energy Efficiency and Renewable Energy Bioenergy Technologies and Vehicle Technologies Offices. Co-Optima is a collaborative project of several national laboratories initiated to simultaneously accelerate the introduction of affordable, scalable, and sustainable biofuels and high-efficiency, low-emission vehicle engines. This work was authored by the National Renewable Energy Laboratory, operated by Alliance for Sustainable Energy, LLC, for the U.S. Department of Energy (DOE) under Contract No. DE-AC36-08GO28308. Funding provided by U.S. Department of Energy Office of Science and National Nuclear Security Administration. The views expressed in the article do not necessarily represent the views of the DOE or the U.S. Government. The U.S. Government retains and the publisher, by accepting the article for publication, ack

–

Publication: Deep reinforcement learning for dynamic control of fuel injection timing in multi-pulse compression ignition engines. IJER (2021).
Multi-fuel injection strategy discovery using deep reinforcement learning in advanced compression ignition engines. IJER (2022).
Machine Learning Ordinary Differential Equations Solver. (2022) in process

Presenters

  • Nicholas T Wimer

    • National Renewable Energy Laboratory

Authors

  • Nicholas T Wimer

    • National Renewable Energy Laboratory

  • Marc T Henry de Frahan

    • National Renewable Energy Laboratory

  • Shashank Yellapantula

    • National Renewable Energy Laboratory

  • Ray Grout

    • National Renewable Energy Laboratory

  • Steven Kiyabu

    • National Renewable Energy Laboratory