Riassunto analitico
Internal combustion engines are machines that convert the thermal energy generated by the oxidation of the fuel into mechanical energy. This process has quite low efficiency and part of the energy is wasted in form of heat, hence a proper cooling system that can dissipate it is needed. In high performance engines, this thermal load and high mechanical loads are superimposed, leading to thermo-mechanical fatigue phenomena. From this point of view, thermal management becomes an aspect that should not be underestimated.
The sizing of cooling circuits is a demanding challenge, as the fluid temperature is closed to saturation and there is the possibility of forming a vapor film that acts as a barrier for heat exchange. Nowadays, the best method for ICE water jacket design is the Computational Fluid Dynamics analysis, in which is possible to consider, not only the dynamics of the fluid but also the possible formation of vapor bubbles or film. Despite the high accuracy, CFD analysis has a big disadvantage in that it requires an elevated computational power, and a long simulation time follows.
In this dissertation, it is developed a methodology that allows, through empirical correlations, to take into account the variation of the fluid-wall heat transfer coefficient, due to the onset of nucleate boiling phenomena, in thermomechanical Finite Element Analysis. In this way, it is possible to estimate the presence of boiling in the early steps of engine head design, when the biphase CFD analyzes are too expensive and too time demanding, obtaining better thermal results in the first structural simulations.
Literature research is made to find the best boiling empirical correlation in terms of accuracy and simplicity. Then, the empirical correlations chosen are applied directly inside the FEA software through user subroutine feature, in particular in this dissertation Dassault Systems Abaqus and MSC Marc are taken into account. This feature is a strong tool that helps the user to solve non-standard problems.
|
