Riassunto analitico
Due to the increasing need to reduce local pollution hot spots and in order to control the phenomena of global warming, there has been growing concern on the use of alternative fuels in the automotive field. According to IEA (International Energy Agency), one fourth of the global CO2 emissions are caused by the transport sector and among those, 75% are caused by road transportation. This has led to the development of new technologies and to the research for technically feasible and economically viable alternative fuels.
Therefore, the focus of this thesis is on a well-known and mature technology, which is the internal combustion engine, running on an innovative type of fuel which is hydrogen, representing a possible solution for the future development of clean engines. To successively develop a control strategy, it is essential to create a model which simulates the actual behavior of the engine accurately, in order to successively develop a control strategy. This has been done by the use of Artificial Neural
Networks (ANN).
This thesis covers the following topics: in Chapter 1, hydrogen properties and the layout of a H2 ICE are presented. Chapter 2 reports a brief analysis of the GT-Power model from which the computations and the simulations were run.
Chapter 3 focuses on the Neural Network-based model, with an emphasis on the modeling of the indicating parameters and showcases the results by looking at each modelled block. Finally, the conclusions highlight possible system developments.
|
