This study aims to develop a model to predict the burst pressure of a dry filament wound cord-rubber composite pressure vessel under hydrostatic internal pressurization using a submodelling based global–local FEA model. The model links the global displacements of a rebar-based model to obtain the local deformation state in a single rhomboidal representative volume. Emphasis is placed on capturing the local stress concentrations in the fibers due to the unique filament winding mosaic pattern. Fiber damage is included in the local model using a maximum principle strain criteria. Verification of the created model is done experimentally on industrially manufactured burst-test specimens. Measurements for displacement during the experiments are taken photographically, while the burst pressure is captured using a pressure transducer. The final error between the burst pressure of the samples and the experimental demonstrators is approximately 6.5%, a marked improvement over conventional models with truss and rebar elements as fibers.

@en

This project aims at developing a model to predict the damage initiation and propagation in a cylindrical filament-wound cord-rubber structure under internal pressurization using non-linear FEA, and is conducted in cooperation with TANIQ BV, Netherlands.
Cord-reinforced rubber composites are used in several safety-critical industries such as oil and gas and civil plumbing. Despite their widespread use, limited research is available that focuses on the single-cycle damage phenomenon that may occur in events such as over-pressurization.

The aim of this project is to close this knowledge gap by developing a theory that accounts for the key damage modes present in CRC structures. This involves experimental studies for material characterization and identification of relevant damage modes, the creation of a novel fibre overlap model that accurately replicates the meso-level filament-wound structure, and translation of the experimentally verified damage modes into functional damage initiation and propagation laws using a global-local FEA model. Verification of the created damage model is done experimentally on samples manufactured and tested at TANIQ, with differences between model predictions and experimental burst pressures being $\approx 6.5\%$. This is a marked improvement over Taniq's current FEA model, which overpredicts the solution by $\approx 27\%$.

The successful implementation of this model would help industries like TANIQ build efficient, strong, and lightweight rubber composite parts for various industries, thus adapting aerospace design principles to the development of more commonplace apparatus.