A parametric study of the numerical characteristics of a coupled CFD algorithm for predicting the dynamics of pneumohydraulic devices
News
A parametric study of the numerical characteristics of a coupled CFD algorithm for predicting the dynamics of pneumohydraulic devices
Abstract
The paper is devoted to the refinement and parametric study of the numerical characteristics of a coupled axisymmetric CFD model of pneumohydraulic device (PHD) dynamics. The model couples a zero-dimensional thermodynamic formulation for the gas cavity and piston motion with the two-dimensional axisymmetric unsteady Navier–Stokes equations for a viscous incompressible working fluid; the implementation is carried out in an in-house code using the finite-volume method on a non-staggered grid with the PISO algorithm. The full statement of the coupled problem is given (geometry, governing equations, initial and boundary conditions, characteristic Reynolds number), and the choice of a laminar flow model is justified. A study of the numerical characteristics of the algorithm is performed: grid convergence (cell size from 3.0 to 0.5 mm) and the influence of the time step, with justification of the baseline configuration (cell size 1.0 mm). The model is verified against experimental data for the steady piston position. It is shown that accounting for the fluid viscosity and compressibility provides a relaxation (damping) effect: the coupled CFD model reproduces the physically justified decay of the transient process, whereas the zero-dimensional model overestimates the oscillation damping time by about ten times. The flow structure (axial velocity profiles and vector fields with recirculation zones) directly demonstrates the damping mechanism. It is concluded that the proposed coupled CFD model is a more reliable basis for predicting the damping characteristics of PHD5, while its in-house implementation ensures the required control over the discretization schemes, boundary conditions and coupling algorithm.
Keywords
Edition
Proceedings of the Institute for System Programming, vol. 38, issue 6, part 2, 2026, pp. 333-346
ISSN 2220-6426 (Online), ISSN 2079-8156 (Print).
DOI: 10.15514/ISPRAS-2026-38(6)-41
For citation
Full text of the paper in pdf (in Russian)
Back to the contents of the volume