Reducing computational time in hydrogen storage simulations

The Faculty of Mechanical Engineering at the Technical University of Košice (TUKE) uses experimental measurements to support modelling and simulation of hydrogen storage tank filling.

Real-world measurements, however, are rarely perfect. The input signals contained fluctuations, isolated extreme values and abrupt changes that increased the complexity of the data entering the simulation.

The challenge was clear:

• Data Quality: Reduce measurement noise while preserving the physical behaviour relevant to the simulation.
• Computational Efficiency: Determine whether better-prepared input data could reduce simulation time without significantly changing the calculated results.

OWASmooth processed selected experimental signals before they entered the simulation. The TUKE research team then performed a controlled benchmark on the PERUN supercomputer, comparing simulations using raw and OWASmooth-processed inputs under comparable computing conditions.

The result showed that improving input data quality can affect not only the data itself, but also the computational resources required to work with it.