Development of heat rejection prediction methodology for selection of cooling elements in diesel engines

Abstract

Internal combustion engines convert chemical energy in fuels such as diesel, gasoline, natural gas, LPG (liquefied petroleum gas) into mechanical energy. The fuel used enters a chemical reaction with the air in the combustion chamber inside the engine and releases heat energy. This heat released increases the gas pressure in the combustion chamber, which causes the piston to move. Engines can be classified according to criteria; such as fuel type, cylinder arrangement, operating time, mixture formation, ignition type (spark ignition - compression ignition), cooling technique (air-cooled, water-cooled), method of filling the cylinder (naturally aspirated, turbocharged, supercharged), or valve arrangement. The most significant environmental and health problems encountered in diesel vehicles are caused by nitrogen oxides and particles emissions. Both of these have very high emissions compared to gasoline vehicles. Emission standards have been established to keep these emissions under control. The amount of NOx (Nitrogen Oxide) formed during combustion is highly rely on temperature. By diluting the mixture in the combustion chamber with the exhaust gases with the help of the EGR (exhaust gas recirculation system), the combustion end temperatures and thus the amount of NOx produced are reduced. The function of this system is to reduce the oxygen concentration in the mixture by sending the exhaust gases back to the cylinders, reducing the mixing ratio, and reducing the maximum gas temperature by raising cylinder gases heat capability . Increasingly stricter emissions regulations are forcing the automotive industry to focus on new technologies ensuring lower emissions. The declaration of the European Union Commission in May 2018 targets 30% lower CO2 (carbon dioxide) emissions compared with 2019 average fleet values in the heavy-duty vehicle market. Conventional diesel engines must operate with maximum energy efficiency to fulfill the requirement. Lowering the engine heat rejection to air and coolants (water, oil) is an obligation to increase energy efficiency and utilize the wasted energy as enhanced exhaust enthalpy. Within the scope of this thesis, the estimation methodology of the total heat rejection to the radiator, which can use in the early stages of an internal combustion engine development, is studied with the help of a 1-dimensional thermodynamic model, and a correlation study is carried out with the test data. It provides a meaningful benefit in the selection of cooling components by accurately estimating the total heat. Furthermore, within the scope of this thesis, the domestic engine (Ecotorq) of F-MAX, a domestic production heavy-duty vehicle belonging to Ford OTOSAN A.Ş., is studied. Ecotorq, analyzed within the scope of this study, is an internal combustion engine with 13L engine displacement and 500PS brake power, 4-stroke, turbocharger and EGR system, Euro6d calibration. The concept first introduced in this study is the coolant and oil temperature measurement procedure to calculate the heat transfer from the combustion chamber walls. The calculated heat transfer is then used as reference data for model correlation. More refined heat transfer reference data is needed separately for the piston, cylinder head, and liner, as the coolant and oil temperature measurement will only give the total heat transfer from the cylinder head, cylinder liner, and piston. Therefore, a 3D-CFD combined heat transfer model, already correlated with the reference test data, is used for the detailed correlation in the combustion chamber sections. Coolant and oil heat values can also be checked by looking at the coolant and oil temperature measurements. Therefore, the heat transfer is calculated from these values. Although the actual heat transfer can obtain with the highest accuracy from the 3D-CFD heat transfer model, it requires an extremely slow and time-consuming process as expected. At this point, the advantage of the thermally correlated 1D engine performance model is that it is really fast, and once the correlation study is complete, it is very reliable for the spatial maximum and average surface temperatures it produces at each operating point for simulations. In the methodology, a 1-dimensional thermodynamic model of the base engine is generated, and a thermal correlation study is carried out to the outputs of the 3-D computational fluid dynamics model, which was previously correlated to the reference test data. During this study, the correlations of critical metal temperature, heat rejection, and exhaust temperature played a vital role. Next, thermal test data is collected in a dynamometer test environment. In addition to the critical performance parameters such as indicated torque and maximum in-cylinder pressure in the model, the total rejected heat from the cylinder &ports, and the results of these test data are at a good correlation level. A separate 1-dimensional model was produced, and a correlation study was performed with the test data to estimate the rejected heat from the EGR cooler. Subsequently, an engine model was created by combining the base engine, EGR cooler model, and turbocharger with the correlation study. An optimization study has been carried out for the "DI-Pulse" combustion model, which is a predictive combustion model. This engine model can predict heat rejection parameters and engine performance parameters at a good level and quickly. Finally, the methodology study was completed by comparing this final thermal model obtained as a result of optimization with the test data collected to compare both engine performance and thermal data. This model can predict total radiator and EGR cooler heat rejection at ±10 kW and ±5 kW for heat dissipated from cylinders and ports. The methodology has been validated by comparing it with test data. With the help of the methodology, unexpected overheating problems can be predicted, the correct design selection of cooling system components can be realized through analytical tools, and cost and time optimization can be achieved by reducing the actual testing needs.