Aerodynamic Design Best Practices Using PowerFlow and modeFRONTIER optimization
The best practices developed by Exa and ESTECO ensure accurate simulation results and a streamlined process for vehicle design optimization
The example in this paper is an aerodynamic guidance study of a generic SUV-type vehicle. The front fascia surface had been frozen, while the rear had high design flexibility. The goal was to understand the impact of rear end design changes to provide guidance for achieving minimized drag, while also considering rear lift and side force fluctuations. The project was conducted on ExaCLOUD, which provided secure, web-based access to the complete Exa simulation software suite and modeFRONTIER. The targeted completion date for the study was two weeks. A bandwidth of four parallel design evaluations was anticipated on ExaCLOUD, with each simulation taking 12 hours. This allowed for 40 simulations to be completed in 5 days, leaving enough time for design space preparation and post-processing.
Even with strict design constraints, the integration of PowerFLOW and modeFRONTIER allowed us to achieve 7.5% drag reduction for one of Liebao's SUVs, which we would not have been able to achieve by hand.
Simulation-driven optimization provides a methodology for achieving aerodynamic performance targets and providing design guidance in a shorter period and with more thorough consideration of design space complexities than a sequential design process.
The PowerFLOW node in modeFRONTIER provides an integrated solution for efficiently and accurately automating aerodynamic design studies. Analysis of the baseline design provides insight for creating design features. Design feature effectiveness is then evaluated by conducting an initial design space characterization. modeFRONTIER intelligent algorithms improve the response surface quality toward overall trend resolution or more targeted design. Optimal designs are predicted using RSM-based optimization and validated with PowerFLOW simulation. Further statistical and flow analysis provide deeper insight to enable flexible, datadriven decision making and recommendations.