Abstract |
The art of injecting the correct amount of fuel
As my master thesis, I have studied a gaseous sequential injection
system (GSI) consisting of a fuel tank, safety shut-off valves, a
vaporiser/pressure regulator (V/P), injectors and an electronic
control module. The aim of this study was to make a mathematical
description of the behaviour of the GSI. The main question has been
how to inject the correct amount of fuel. It was important that the
developed models could be transformed into real-time running
algorithms. As the GSI injectors were still under development, no
study of the injectors has been done. The work has instead been
focused on determining the parameters that affect the actual injected
amount of fuel, and how to derive them from the parameters measured by
the available sensors. Steady state models describing the pressures and outlet
temperature of the vaporiser/pressure regulator, the temperature
change of the fuel between the V/P and the injectors, outdoor
temperature and tank temperature, have been made. The pressure models
have been made by identification, while the models for the outlet fuel
temperature of the V/P and the change in temperature before the
injectors are based on physical knowledge. The models are all
describing the measured data well. Transient models have been made for the outlet temperature of the
V/P and the change in fuel temperature between the V/P and the
injectors. When making these models, emphasis was put on making the
algorithms easy to calculate in the electronic control module. The
transient behaviour has thus been approximated with first order
filters and offsets, which gave rather good results. The developed models can be transformed into real-time running
algorithms. There are some things left to be described, but it seems to be
possible to make a real-time running software which calculates how to
inject the correct amount of fuel from the information given by the
available sensors.
Ylva Nilsson
1999


Page responsible: webmaster
Last updated: 2021-11-10