Date of Award


Publication Type

Master Thesis

Degree Name



Mechanical, Automotive, and Materials Engineering


Defoe, Jeff




Currently, full vehicle computational fluid dynamics (CFD) simulations are used to predict rear fascia temperatures. As these simulations are expensive and time consuming, only what is intended to be a worst case scenario analysis is completed. Certain variables can be overlooked and the case selected may not be the worst case scenario. The objective of this thesis is to create a surrogate model that can rapidly predict maximum fascia temperature for a variety of vehicle operating conditions and exhaust positions, while exploring the physical mechanisms responsible for the heat transfer between the exhaust gas, exhaust components, and rear fascia. Using full vehicle CFD simulations, an investigation of the maximum fascia temperature as a function of vehicle operating conditions and exhaust positioning is completed by identifying non-dimensional parameters governing maximum fascia temperature, consisting of both geometric and non-geometric parameters based on vehicle speed and exhaust inlet velocity (Reynolds number), their ratio (velocity ratio), and exhaust temperature (exhaust temperature ratio). The exhaust positioning within the rear fascia is simplified into four non-dimensional parameters to explore the modifications in geometry. A design of experiments (DOE) was completed with full vehicle CFD simulations using optimal Latin hypercube sampling of the input variables. Using data from the DOE, a surrogate model is generated. The individual impact of each parameter on the maximum fascia temperature is identified and the surrogate model suggests a vehicle operating condition consisting of low vehicle speed and high load (high exhaust velocity and exhaust gas temperature) results in highest fascia temperatures. For this condition, at a baseline exhaust position, the maximum fascia temperature exceeds the maximum allowable value by 200 K. For the same operating condition, the exhaust positioning predicted by the surrogate model to result in lowest maximum fascia temperature exceeds the maximum allowable value by only 70 K.