Application of modified k-ω model to predicting cavitating flow in centrifugal pump

Considering the compressibility of the cavity in the cavitating flow, this paper presents a modified k-ω model for predicting the cavitating flow in a centrifugal pump, in which the modified k-ω model and Schnerr-Sauer cavitation model were combined with ANSYS CFX. To evaluate the modified and standard k-co models, numerical simulations were performed with these two models, respectively, and the calculation results were compared with the experimental data. Numerical simulations were executed with three different values of the flow coefficient, and the simulation results of the modified k-ω model showed agreement with most of the experimental data. The cavitating flow in the centrifugal pump obtained by the modified k-ω model at the design flow coefficient of 0.102, was analyzed. When the cavitation number decreases, the cavity initially generates on the suction side of the blade near the leading edge and then expands to the outlet of the impeller, and the decrease of the total pressure coefficient mainly occurs upstream of the impeller passage, while the downstream remains almost unaffected by the development of cavitation.

A high precision computing method for heat transfer in the process of oil-water displacement

The process of oil-water displacement is one of the key technologies in offshore underwater tank.When hot oil is contacting the normal-temperature water,the interfacial heat transfer should be investigated as the heat loss may result in wax precipitation and solidification which will reduce the flowing of oil and thus affect the process.As for the numerical simulation of heat transfer,the calculation is costly as the underwater tank is usually large and the displacement period is long.A high precision computing method would greatly reduce the mesh scale.Therefore,this research is performed to establish a high precision computing solver.Based on volume of fluid(VOF),a new form of energy equation is proposed.This equation is derived from temperature equation and the variable internal energy per volume is used.This variable is additive and has a close relationship with volume fraction.With algorithm implantation to OpenFOAM,two non-isothermal VOF solvers are established corresponding to temperature equation and the new equation respectively.After an analytical solution is built,the two solvers are compared.The solver based on the new equation presents far more accurate results than the solver based on temperature equation.An energy weighted scheme is more reasonable than a linear temperature distribution for the mixture phase.