Finite element bicycle helmet models development

Abstract Impact attenuation performance of three different range of commercial bicycle helmet were investigated in lateral drop impact test in accordance to AS/NZS 2063:2008, Australian/New Zealand Standard for bicycle helmet using numerical simulation and and experimental impact test. The aim of this research is to develop a simulation model of drop impact test, which to be used in further investigations of user-centred design approach of bicycle helmet. Three commercial bicycle helmet models were used in this study. All helmets and J headform were scanned using Flexscan 3D scanning equipment. Post-scan processing jobs of scanned geometry models such as helmet liner, shell and headform were conducted in Geomagic Studio 12. The experimental impact test is carried out using 2-wire drop test facility in accordance to the AS/NZS 2063:2008, Australian Standard for bicycle helmet. A few samples were cut from the liner of each helmet to determine the density of Expanded Polystyrene (EPS). Headform peak linear acceleration, impact duration and impact speed of each helmet were measured and recorded from the drop test. The scanned geometry models were imported into Abaqus. A drop impact simulation was developed based on the density and impact speed data obtained from the physical test. Inner liner of bicycle helmet, made from Expanded Polystyrene (EPS), was modeled using crushable foam properties, while headform and anvil were modeled as rigid bodies. Peak linear accelerations and impact duration of the headform on each helmet at three different impact locations of helmet were recorded. A robust correlation study using peak linear acceleration score, impact duration score and Pearson correlation coefficient between the data from physical test and numerical model was conducted. Good correlation scores (>80%) were achieved between the numerical model and experimental impact test in terms of headform peak linear acceleration and impact duration score, suggesting that the simulation model is in good correlation with those from physical test.