Study On The Constitutive Model And Large Eddy Simulation Of Drag-reducing Flow With Surfactant

Turbulent drag reduction(DR)by additives is an important technology for energy saving of long-distance pipeline transportation of oil and heating pipeline system in oil field.It can reduce pumping energy or improve flow rate.Although surfactant possesses significant advantages and promising application prospect,the mechanism of turbulent DR induced by surfactant is still not well understood,especially the current direct numerical simulation of turbulent DR is limited to the flow with low Reynolds number,mechanism of turbulent DR in high Reynolds number is not studied deeply and widely,which hinders its industrial application.To promote the application of turbulent DR technology in oil industry,numerical and experimental methods are carried out in this dissertation to study on the constitutive model and large eddy simulation(LES)of drag-reducing flow with surfactant,and the mechanism of turbulent DR with high Reynolds number is further studied.For constitutive model,the anisotropy in relaxations and deformations of microstructures formed in the surfactant solution is considered and the thought of multiple relaxation times is taken into account,based on which an N-parallel FENE-P constitutive model is put forward.It aims at describing the rheological properties of surfactant solution more accurately.Comparative results indicate that the proposed N-parallel FENE-P constitutive model can describe the apparent viscosity and first normal stress difference of surfactant solutions more accurate than the traditional FENE-P constitutive model which characterized with single relaxation time.In addition,the N-parallel FENE-P constitutive model demonstrates a better applicability as well as favorable adjustability of the model parameters.For large eddy simulation,(1)an improved coherent-structures Smagorinsky model(ICSM)in which the energy-decay suppression function is redefined from the perspective of physical meaning is proposed.Furthermore,an improved mixed subgrid-scale(SGS)model which combines ICSM with temporal approximate deconvolution model(TADM)named MICT is established for the LES of surfactant-induced DR in turbulent channel flow.Simulation results demonstrate that,compared with the conventional MCT SGS model,results obtained by MICT agree better with those of DNS and the calculation accuracy of the LES of turbulent DR with surfactant is improved.(2)Based on N-parallel FENE-P constitutive model and improved mixed SGS model(MICT),the governing equation of LES of turbulent drag-reducing channel flow with surfactant is established first.The LES governing equation is a complicated system with nonlinear and multi-variables coupled characteristics.To solve the LES governing equation efficiently and accurately,the selection of convective discretization scheme for constitutive model,setting of restriction operator for multigrid method as well as convergence criterion for discretized LES governing equation are studied in detail in this dissertation.Based on above studies,an in-house LES code for the calculation of drag-reducing channel flow with surfactant additive is developed using FORTRAN language.The code is validated by PIV data of turbulent drag-reducing channel flow.Then the turbulent drag-reducing channel flow with high Reynolds numbers is investigated,the simulation results including turbulent DR rate,representative turbulent statistical data are analyzed.Furthermore,the contributions of turbulent frictional resistance coefficient,balance of different parts to turbulent kinetic energy,and quadrant distribution of Reynolds shear stress as well as coherent structures closely relating to turbulent burst event are deeply studied and analyzed.The above studies can provide theoretical guidance for the application of turbulent DR technology on long-distance pipeline transportation of oil and heating pipeline system in oil field.