Authors:
Vijay Manikandan Janakiraman
1
;
Jeff Sterniak
2
and
Dennis Assanis
3
Affiliations:
1
University of Michigan, United States
;
2
Robert Bosch LLC, United States
;
3
Stony Brook University, United States
Keyword(s):
Support Vector, Identification, Combustion, Homogeneous Charge Compression Ignition, HCCI, Neural Networks, Nonlinear Regression, Engine Model, Control Model.
Related
Ontology
Subjects/Areas/Topics:
Engineering Applications
;
Informatics in Control, Automation and Robotics
;
Intelligent Control Systems and Optimization
;
Machine Learning in Control Applications
;
Robotics and Automation
;
Signal Processing, Sensors, Systems Modeling and Control
;
System Identification
;
System Modeling
Abstract:
Homogeneous charge compression ignition (HCCI) is a promising technology for Internal Combustion Engines
to improve efficiency and reduce nitrogen oxides emissions. Control of HCCI combustion is often
model-based, and it is vital to have a good model of the engine to make control decisions. The HCCI engine is
characterized by complex chemical kinetics whose physical modeling is difficult and laborious. Identification
is an effective alternative to quickly develop control oriented models for such systems. This paper formulates
a Support Vector Regression (SVR) methodology for developing identification models capturing HCCI combustion
behavior. Measurable quantities from the engine such as net mean effective pressure (NMEP) and
crank angle at 50% mass fraction burned (CA50) can be used to characterize and control the HCCI engine and
are considered for identification in this study. The selected input variables include injected fuel mass (FM)
and valve events {intake valve opening (IVO),
exhaust valve closing (EVC)}. Transient data from a gasoline
HCCI engine recorded at stable HCCI conditions is used for training, validating and testing the SVR models.
Comparisons with the experimental results show that SVR with Gaussian kernels can be a powerful approach
for identification of a complex combustion system like the HCCI engine.
(More)