Scaling analysis as a tool to validate CFD simulation of a lubricant flow in the bearing housing of a gas turbine

Abstract

• The effects of rotor speed on the temperature of the lubricant oil are studied. • CFD model of oil flow and heat transfer in the bearing chamber is developed. • Scaling analysis is used to simplify and nondimensionalize the governing equations. • Scaling analysis is used to replace experimental data to validate CFD results. • An equation to compute viscous dissipation energy is obtained by Scaling analysis. This work concentrates on the journal bearing chamber of a gas turbine engine, to explain the impacts of the rotational speed of the rotor on the velocity, pressure and temperature distribution in the bulk oil. In the first step, a CFD model of fluid flow and heat transfer in the bulk oil of the bearing chamber was developed via COMSOL 6. Next, and in a novel approach, scaling analysis was employed to verify the CFD simulation results. Scaling analysis, as a systematic method for non-dimensionalizing a set of equations, was used to simplify the governing equations based on the specific conditions of the physics. Afterward, the velocity, pressure and temperature profiles attained from the CFD simulation were compared to this set of simplified equations. In general, it was confirmed that the CFD model developed in the current article is physically valid and follows the main descriptive equations governing the entire physics of the system. Scaling analysis was effectively used to confirm the simulation results where insufficient experimental results were available, due to the geometrical and operational restrictions in the test rig. Temperature, pressure and velocity were the factors considered in this study and the average of simulation and scaling results for each factor were compared. Additionally, the average point-to-point error between the simulation and scaling results for each factor was reported as another basis for the comparison. In summary, scaling analysis in this work is presented as a straightforward method to substitute experimental measurements in order to validate CFD simulation results.