Abstract

Type IV composite pressure vessels represent the current state-of-the-art for compressed gaseous hydrogen storage in fuel cell electric vehicles. A combination of highly demanding safety regulations and the need for cost competitive solutions make the topic of CPVs particularly challenging. Given the elevated material price of carbon fiber, structural optimality is essential to meet both requirements. Thorough understanding of design parameters and mechanical performance of composite pressure vessels is prerequisite to structural optimality. In this paper we investigate the relation of stacking sequence and circumferential ply drop locations on the mechanical performance of type IV composite pressure vessels subjected with internal pressure. This paper builds on previous studies by the authors, which are enhanced by new numerical and experimental results. An experimental set is used, where for a given layup composition the stacking sequence and the circumferential ply drop locations are varied. The experimental results are complemented by a computationally efficient numerical framework, which is composed by the output of a commercial filament winding software, a self-developed geometry correction algorithm and an automated FE model generation program. The numerical results are compared with outer surface strains obtained by means of three-dimensional digital image correlation, the final burst pressures and the vessel remainders. The achieved burst pressures vary between 152.6 and 188.6MPa, depending on design configuration. For the layup composition in this investigation, the placement of tangentially reinforcing layers (e.g. circumferential and high-angle helical layers) as innermost layers led to overall higher cylinder strengths compared to sequences, where these layers were located as outermost. The retraction of circumferential ply drop locations was found to impact the burst performance differently in dependence of the stacking sequence. For sequences, where circumferential layers were located as outermost, a retraction of ply drop locations by 12mm showed barely any differences in burst pressure (-1.9%). For sequences, where these layers were located as innermost, a severe decrease (-19.1%) was noticed once the ply drop locations were retracted by up to 9mm. The results not only underlined the criticality of both design parameters and their interaction with each other, but also showcased a computationally efficient numerical framework capable of capturing distinct mechanical responses for a variety of layups at least trend-wise.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call