Five models and methodology are discussed in this paper for constructing classifiers capable of recognizing in real time the type of fuel injected into a diesel engine cylinder to accuracy acceptable in practical technical applications. Experimental research was carried out on the dynamic engine test facility. The signal of in-cylinder and in-injection line pressure in an internal combustion engine powered by mineral fuel, biodiesel or blends of these two fuel types was evaluated using the vibro-acoustic method. Computational intelligence methods such as classification trees, particle swarm optimization and random forest were applied.
Experimental investigations and numerical simulations have been conducted in this study to derive and test the values of kinetic parameters describing oxidation and gasification reactions between char carbon and O2 and CO2 occurring at standard air and oxy-fuel combustion conditions. Experiments were carried out in an electrically heated drop-tube at heating rates comparable to fullscale pulverized fuel combustion chambers. Values of the kinetic parameters, obtained by minimization of the difference between the experimental and modeled values of char burnout, have been derived and CFD simulations reproducing the experimental conditions of the drop tube furnace confirmed proper agreement between numerical and experimental char burnout.
The aim of this work was to investigate the heat and mass transfer during thermal decomposition of a single solid fuel particle. The problem regards the pyrolysis process which occurs in the absence of oxygen in the first stage of fuel oxidation. Moreover, the mass transfer during heating of the solid fuels is the basic phenomenon in the pyrolysis-derived alternative fuels (gas, liquid and solid phase) and in the gasification process which is focused on the generation of syngas (gas phase) and char (solid phase). Numerical simulations concern pyrolysis process of a single solid particle which occurs as a consequence of the particle temperature increase. The research was aimed at an analysis of the influence of particle physical properties on the devolatilization process. In the mathematical modeling the fuel grain is treated as an ideal sphere which consists of porous material (solid and gaseous phase), so as to simplify the final form of the partial differential equations. Assumption that the physical properties change only in the radial direction, reduces the partial derivatives of the angular coordinates. This leads to obtaining the equations which are only the functions of the radial coordinate. The model consists of the mass, momentum and energy equations for porous spherical solid particle heated by the stream of hot gas. The mass source term was determined in the wide range of the temperature according to the experimental data. The devolatilization rate was defined by the Arrhenius formula. The results of numerical simulation show that the heating and devolatilization time strongly depend on the physical properties of fuel. Moreover, proposed model allows to determine the pyrolysis process direction, which is limited by the equilibrium state.