基于数据同化的液环真空泵内部流动分析OA

To address the significant deviation between numerical simulation and experimental test results caused by the complex gas-liquid two-phase flow in liquid ring pumps,and to overcome the limitations in the number and locations of experimental measurement points,the ensemble Kalman filter(EnKF)method was employed to assimilate the flow field of the liquid ring pump based on numerical simulation and limited experimental test data.Latin hypercube sampling was used to generate combined samples of the turbulence coefficients in the Renormalization Group(RNG)k-ε model.Assimilation analysis of the turbulence model coefficients was performed based on pressure data from four measurement points on the inner wall of the pump casing.The error between the assimilated results and the experimental observations is 2.77％,which represents a reduction of 1.53％compared to the original computational fluid dynamics(CFD)prediction error.A comparison of the post-assimilation flow field characteristics show that constrained by the experimental values on the inner wall of the casing,the overall pressure inside the pump is increased to some extent.Compared with the original CFD results,the influence of the assimilated phase field is mainly observed in the gas-liquid interface region.After assimilation,the turbulent viscosity coefficient Cμ decreases,leading to a reduction in local momentum dissipation and a corresponding increase in the velocity distribution.Meanwhile,the decrease in the turbulence dissipation rate coefficient C1ε,and the increase in C2ε result in a reduction of turbulence dissipation rate and an increase in turbulent kinetic energy.