Multi-parametric optimization of aerodynamic performance and pedestrian crash in sedan front-end profiles

Enhancing vehicle front-end (VFE) design is critical to improving aerodynamic efficiency and pedestrian safety. However, most existing designs focus primarily on reducing the drag coefficient (Cd) to meet aerodynamic standards, often overlooking the Head Injury Criterion (HIC), which is essential in minimizing pedestrian injuries. According to the Malaysia Road Fatalities Index, nearly 9% of fatal road accidents between 2010 and 2019 involved pedestrians. Despite the addition of safety features such as airbags and intelligent speed assist (ISA), these solutions have not always achieved their intended effectiveness and may compromise overall vehicle performance. While previous studies have examined either aerodynamic or safety aspects of VFE designs, few have conducted a comprehensive parametric optimization addressing both objectives simultaneously. Therefore, this study aims to optimize the front-end design of sedan vehicles to achieve a balance between low aerodynamic drag and enhanced pedestrian protection, particularly during low-speed impacts. A multiobjective optimization framework was developed using computational simulations, mathematical modelling, and evolutionary algorithms. The sedan vehicle models were initially designed using CATIA V5R21 and subsequently imported into ANSYS 2024 Rl for aerodynamic analysis. The model was enclosed in a fluid domain replicating wind tunnel conditions, maintaining conventional dimensions of three car lengths (12,600 mm) in front, behind, above, and beside the vehicle. Simulations were conducted using the realizable k-e turbulence model in ANSYS Fluent to accurately capture external airflow behaviour. A mesh convergence study was performed on four sedan profiles by initially employing smaller element sizes to ensure solution accuracy. The element size was gradually increased to identify the maximum allowable size in the critical airflow impact regions while maintaining consistent results. Through this process, it was determined that a 500 mm element size represents the maximum size that still offers an effective balance between computational cost and analysis accuracy, confirming mesh convergence for subsequent simulations. The Central Composite Design (CCD) method was employed to investigate the impact of seven critical VFE parameters, including windshield angle and hood edge height, on Cd and HIC values. MATLAB's Model-Based Calibration (MBC) toolbox was used to generate response surface models, achieving an RMSE of 0.0065 and R2 of 89.2% for Cd, and an RMSE of 0.01 and R2 of 81.7% for HIC. Genetic Algorithms (GA) were used to perform multiobjective optimization, yielding an optimal design with a drag coefficient (Cd) of 0.1908 and a Head Injury Criterion (HIC) of 99.7241. The design was validated through CFD simulation in ANSYS, producing a simulated Cd of 0.2132, with an acceptable 11.74%) deviation from the optimized value, remaining within engineering tolerance limits. This thesis highlights the significant influence of parameters such as bumper centre height and hood edge height on both aerodynamic and pedestrian safety performance. The proposed framework offers a robust approach for designing front-end vehicle geometries that align with modern safety regulations while enhancing fuel efficiency. These findings provide valuable guidance for the automotive industry in developing vehicles that are both high-performing and pedestrian-friendly.