Date of Award
Mechanical, Automotive, and Materials Engineering
1D-3D model integration, Exhaust skin temperature, Exhaust system heat transfer, External emissivity, K-means clustering, Vehicle level CFD
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.
Studies involving flow and heat transfer in automotive exhaust systems are regularly employed in the design and optimization phases. Both internal as well as external heat transfer are key to provide a better understanding of the underbody heat transfer, cold start warm-up and thermal aging of the catalytic converter for gasoline engines and adequate thermal protection for the underbody components. The internal flow in a typical automobile exhaust system can be simplified using a 1D model employing correctional factors to encompass the three-dimensional effects. However, the external flow and heat transfer underbody of a vehicle is highly complex as it involves the overall front-end design of the car as well as the packaging of components underhood and underbody. This would require the use of a full scale 3D model of a vehicle. The proposed research involves the prediction of exhaust skin (outer surface) temperature combining a 1D model with a full vehicle 3D model as well as investigating heat transfer characteristics of the exhaust system. The 1D model is developed using a commercial code, GT-Power and the 3D vehicle level model is simulated using STAR-CCM+. The 1D and the 3Dmodel will provide a real time closed loop control system based on the combustion requirements and exhaust system readings for internal flow and external flow. In the first stage, the gas side internal heat transfer is simulated using the 1D model by adding available heat transfer correlations considering entrance effects, engine induced pulsation, geometrical effects and surface conditions. Initially, the model is simulated for steady state wide open throttle (WOT) cases and validated with results available from bench test. In the second stage, the use of the model is extended further in transient heat transfer studies. In the third stage, the 3D vehicle level model is simulated using the commercial code STAR-CCM+ at various wind speeds based on a set of cluster points representing a transient drive cycle. A Reynold Averaged Navier-Stokes (RANS) based k-ε turbulence model is used for modeling flow and turbulence. Thermal models for free convection and thermal radiation, are used to account for external heat transfer. The initial thermal boundary condition of the exhaust for the simulation is obtained from the preliminary 1D simulation data. The predicted external heat transfer coefficients from the 3D model are then used as a boundary condition for the 1D model for heat transfer as a third phase of the study. The iterative of the process of using the 3D model as boundary condition for the 1D model and vice versa until convergence will ensure a more accurate prediction of the exhaust skin temperature. Further a parametric study involving the influence of external emissivity on exhaust system heat transfer was carried out. The results indicate that the effect of the external emissivity is significant on the skin temperature and external heat transfer. The variation in emissivity is seen to contribute to more than 50% in the overall heat transfer. A temperature difference of up to 200oC was seen on the heat shields of the exhaust at high loads. Similar results were seen for the other components underbody close to the exhaust system. This would potentially be higher at idling after a drive cycle where free convection and radiation are seen to be more dominant, indicating a strong influence of external radiation as a key parameter in the heat transfer from an exhaust. Further the study revealed that the variation in emissivity does not influence the convective heat transfer by more than 4%.
Annur Balasubramanian, Sudharsan, "Prediction of Exhaust Skin Temperature Integrating 1D Model with Vehicle Level CFD Model" (2018). Electronic Theses and Dissertations. 7599.