SVILUPPO DI UN MODELLO FENOMENOLOGICO MULTIZONA PER L'ANALISI DEL PROCESSO DI COMBUSTIONE NEI MOTORI DIESEL

Abstract

This research project was developed starting from the aim to combine the performance improvement with observance of the emission limits in the in the direct-injection diesel engines. These considerations show, therefore, the need to investigate the diesel spray. Based on this assumption, it was in fact developed a new descriptive model of these phenomena which formed the input for the development of innovative calculation models for numerically simulate combustion and pollutant formation. By analyzing the experimental observations of John Dec, it was developed and implemented a new phenomenological multizone model. Before describing the combustion process, it is provided for the characterization of the spray as reported by Dennis Siebers. Subsequently, the diesel jet has been divided into six zones, solved so that to obtain information relating to the medium pressure in the cylinder, to the heat release curve, to the masses and temperatures. The description of the six zones is based on some sub models for the representation of the injection law, the Siebers model for the diesel spray, the hot air entrainment in the spray, the Watson predictive model for premixed and diffusive combustion and the energy balance in each area with its mean temperatures. In order to characterize the model, it has been hypothesized that the single jets do not interact with each other and behave in the same way, there is no interaction between the jet and the walls of the combustion chamber, the pressure is uniform, temperatures and composition are variable from zone to zone, shape and size of the jet will remain the same for each injection nozzle and cycle, there are no distortion effects for swirl / swish and, around the TDC, the thermodynamic conditions in the cylinder during the injection are kept constant. Furthermore, masses and temperatures have been obtained and then it was possible to develop a simple model of soot formation and oxidation. The devised model in MATLAB environment is purely predictive and this constitutes its originality: it allows to accurately follow the trends reported in the literature and make predictions compared to engine performance and pollutant emissions.