Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Learning Based Model Predictive Control of Combustion Timing in Multi-Cylinder Partially Premixed Combustion Engine

Li, Xiufei LU ; Yin, Lianhao LU ; Tunestål, Per LU and Johansson, Rolf LU orcid (2019) In SAE Technical Papers
Abstract
Partially Premixed Combustion (PPC) has shown to be a promising advanced combustion mode for future engines in terms of efficiency and emission levels. The combustion timing should be suitably phased to realize high efficiency. However, a simple constant model based predictive controller is not sufficient for controlling the combustion during transient operation. This article proposed one learning based model predictive control (LBMPC) approach to achieve controllability and feasibility. A learning model was developed to capture combustion variation. Since PPC engines could have unacceptably high pressure-rise rates at different operation points, triple injection is applied as a solvent, with the use of two pilot fuel injections. The LBMPC... (More)
Partially Premixed Combustion (PPC) has shown to be a promising advanced combustion mode for future engines in terms of efficiency and emission levels. The combustion timing should be suitably phased to realize high efficiency. However, a simple constant model based predictive controller is not sufficient for controlling the combustion during transient operation. This article proposed one learning based model predictive control (LBMPC) approach to achieve controllability and feasibility. A learning model was developed to capture combustion variation. Since PPC engines could have unacceptably high pressure-rise rates at different operation points, triple injection is applied as a solvent, with the use of two pilot fuel injections. The LBMPC controller utilizes the main injection timing to manage the combustion timing. The cylinder pressure is used as the combustion feedback. The method is validated in a multi-cylinder heavy-duty PPC engine for transient control. (Less)
Please use this url to cite or link to this publication:
author
; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
in
SAE Technical Papers
article number
Technical Paper 2019-24-0016
publisher
Society of Automotive Engineers
external identifiers
  • scopus:85085864235
ISSN
0148-7191
DOI
10.4271/2019-24-0016
project
Competence Centre for Combustion Processes
language
English
LU publication?
yes
id
3c67f93d-44b6-46f8-8747-7df6fe3783a8
date added to LUP
2020-03-12 16:36:16
date last changed
2022-04-18 21:29:30
@article{3c67f93d-44b6-46f8-8747-7df6fe3783a8,
  abstract     = {{Partially Premixed Combustion (PPC) has shown to be a promising advanced combustion mode for future engines in terms of efficiency and emission levels. The combustion timing should be suitably phased to realize high efficiency. However, a simple constant model based predictive controller is not sufficient for controlling the combustion during transient operation. This article proposed one learning based model predictive control (LBMPC) approach to achieve controllability and feasibility. A learning model was developed to capture combustion variation. Since PPC engines could have unacceptably high pressure-rise rates at different operation points, triple injection is applied as a solvent, with the use of two pilot fuel injections. The LBMPC controller utilizes the main injection timing to manage the combustion timing. The cylinder pressure is used as the combustion feedback. The method is validated in a multi-cylinder heavy-duty PPC engine for transient control.}},
  author       = {{Li, Xiufei and Yin, Lianhao and Tunestål, Per and Johansson, Rolf}},
  issn         = {{0148-7191}},
  language     = {{eng}},
  publisher    = {{Society of Automotive Engineers}},
  series       = {{SAE Technical Papers}},
  title        = {{Learning Based Model Predictive Control of Combustion Timing in Multi-Cylinder Partially Premixed Combustion Engine}},
  url          = {{http://dx.doi.org/10.4271/2019-24-0016}},
  doi          = {{10.4271/2019-24-0016}},
  year         = {{2019}},
}