b'Automotive | Engineer InnovationCost of solving Optimization of an AutomotiveissuesThermal Management System Cost of implementing changesBy Fabrcio Thomaz, FCA Fiat Chrysler Automobiles, Cssio Chamone, Pontifcia Universidade Catlica de Minas Gerais, Gustavo Maia, Pontifcia Ability to make Universidade Catlica de Minas Gerais, & Guilherme Tondello, Creative Solutions.changes Figure 1: Risk controllability, development costs, and effort to eliminate an error during Concept Design Production Maintenance project development stagesFlomaster there is also is the ability to use Design of Experiments (DOE) to optimize component parameters virtually. It is important that the cooling system can meet the requirements for extreme vehicle operation which represent the worst case scenarios even though these conditions are encountered less than 1% of the time. To get a good representation of different driving conditions seven scenarios are captured as shown in Table 1.Cooling system modelAccurately capturing the behavior ofFigure 3: Simcenter Flomaster Cooling Modelthe vehicle virtually is often a challenge, for this study four mainareas were the focus: engine heatOnce these values were determined it rejection, fan modeling, radiatorwas very straightforward to look up the102 103modeling, and the overall systemheat rejection from the engine thermal configuration. The most important ofcharacterization map (Figure 2), which these is properly determining thewas previously available. 87 86 86engine heat rejection since this has a84direct impact on the requiredThe next area to model was the fan and performance of the thermalradiator, these were important because75 76management system. The first step wasthey are a major part of removing the to obtain data for the vehicle, inwaste heat from the engine but also particular, engine type, weight, gearbecause these are the main ratio, coast down, tires, final drive ratio,components for optimization in this towing, and payload. This informationstudy. Modeling of the fan is allowed engine speed and torque to bestraightforward using pressure rise calculated for each of the test scenarios.versus volumetric flow rate data along Engine speed was determined from thewith fan affinity laws. These laws allow gear ratio, final drive ratio, tires andthe fan to be scaled virtually in the vehicle speed, while the torque wasmodel based on the original data. A determined by calculating thesimilar approach was used for the resistance to movement using the coastradiator model with a Nusselt Number down and calculating the power toversus Reynolds Number for the coolant overcome obstacles for each testand airside surface map. Because this scenario. It is also important to notemap is dimensionless, it allows the test that when calculating the power to thedata to be scaled based on size, fluids,Experimental Virtualengine, a drivetrain loss of 8% and anand other variations without changing accessory load of 5 hp was added in.the performance map. Figure 4: Model validation of temperatures25'