The 10th Asian Symposium on Computational Heat Transfer and Fluid Flow ( ASCHT2025)

In the context of long-distance product oil pipelines, 1Dmodels can predict only the length of the mixed oil zone and the approximate axial concentration. In contrast, traditional high-dimensional full-pipeline calculations are both extremely costly and beyond the capabilities of FLUENT. To address the trade-off between accuracy and efficiency, this paper proposes a method based on sliding mapping of mixed oil zones. In the context of the FLUENT solver, the use of user-defined functions (UDFs) to construct low-cost computational domains enables the accurate simulation of mixed oil segment formation and high-dimensional concentration distributions within the framework of long-distance finished oil pipelines. The outcomes of this simulation demonstrate a high degree of concordance between the simulated length of the mixed oil segment and the Austin empirical formula. Furthermore, this method effectively captures the trailing phenomenon [1] of mixed oil segments caused by high-viscosity oil flowing behind. This approach has been demonstrated to markedly reduce the computational expense associated with accurately simulating mixed oil concentration distributions in extensive pipelines utilizing the FLUENT solver.
 

sliding mapping, long-distance, low-cost, accurate simulation, trailing phenomenon