A computational fluid dynamic-based method for analyzing the nonlinear relationship between windage loss and pressure in a geotechnical centrifuge

Temperature control is an important limitation to further increase in geotechnical centrifuge power. Although vacuum pumps can reduce windage loss, they also negatively affect heat transfer performance. Therefore, in this study, we aim to accurately determine the rate at which windage loss decreases with pressure to help assess whether reducing pressure is beneficial to temperature control. A computational fluid dynamic method based on the multi-reference model and k–ω shear-stress transport turbulence model is used to simulate the ZJU400gt geotechnical centrifuge. The windage loss and temperature of ZJU400 at 0–150 gravity acceleration under normal pressure conditions are simulated. Compared with the experimental data, the error is < 20.7%, indicating simulation reliability. Furthermore, the simulation model is used to simulate the windage loss power under low-pressure conditions and predict the relationship between the windage loss power and pressure. Compared with current calculation methods, which yield a linear relationship between windage loss and operating pressure, the simulation results indicate a slightly nonlinear relationship. At 5,000 Pa, the simulated windage loss is 40% larger than the calculated value, severely affecting the temperature control design. Moreover, the velocity exhibits minimal variation with pressure, whereas the effective kinematic viscosity varies substantially. The nonlinear relationship between the windage loss and pressure can be attributed to increased turbulent kinetic energy and the size of the wake region caused by vacuum pumping. A formula for nonlinear windage loss with pressure is proposed, providing a basis for the future design of super-gravity geotechnical centrifuges.