Performance of the fractional step method with various temporal discretization and adaptive time step size schemes for pulsating ventilation

Pulsating ventilation has been drawing extensive attention recently. Computational fluid dynamics (CFD), as a widely used and effective tool for investigating pulsating ventilation, often consumes significant computation time. To identify a suitable numerical scheme for this circumstance, we adopted the standard incremental pressure-correction (SIPC) method with higher-order temporal discretization schemes to simulate indoor airflow. To further improve the simulation efficiency, two adaptive time step size schemes were proposed and used to simulate both long-period and short-period pulsating ventilation conditions. Results showed that the SIPC scheme offers accuracy comparable to the PISO (pressure-implicit with splitting of operators) algorithm while saving about 40% of computation time. Higher-order temporal discretization schemes have minimal impact on the accuracy and stability of the SIPC scheme for simulating pulsating airflow, with the first-order Euler backward implicit scheme showing slightly higher efficiency. Compared to the conventional fixed time step size scheme (fixed scheme), both adaptive time step size schemes significantly reduce computation time with negligible impact on accuracy. The scheme that controls time step size based on a given maximum Courant number (MaxCo scheme) saves about 35% of computation time, while the scheme that combines a given maximum Courant number with the curvature of the inlet velocity-time curve (MaxCo+K scheme) to control time step size saves nearly 30%. Although the MaxCo+K scheme requires about 10% more computation time than the MaxCo scheme, it improved accuracy by approximately 10% by more accurately capturing the inlet velocity boundary condition in the short-period pulsating ventilation simulation.