Entity

Time filter

Source Type


Xu Y.,Key Laboratory of Process Measurement and Control in Tianjin | Xu Y.,Tianjin University | Gao L.,Key Laboratory of Process Measurement and Control in Tianjin | Gao L.,Tianjin University | And 3 more authors.
Flow Measurement and Instrumentation | Year: 2014

Based on the operational conditions of the PetroChina Southwest Oil & Gas well field, this study aims to explore the wet gas flow overreading (OR) characteristics of a nonstandard long-throat Venturi by the means of computational fluid dynamics (CFD) technique. The studied prototype structure size is an inner diameter of 50. mm, a diameter ratio of 0.4 and a throat length of 50. mm. According to the field experiment, the simulation pressure is 3. MPa gauge. Through a comparative study of the multiphase flow models and turbulence models, combined with the analysis of the Baker's flow regime and interparticle space under the field conditions, this paper eventually employed DPM model and Eulerian model for wet gas simulation, respectively, and RSM for turbulence model. An equivalent droplet diameter adjustment method was implemented to improve the precision of prediction. During post-processing, the liquid phase distributions and the wall pressure profiles were investigated. The numerical results indicate that the differential pressure in convergent section of long-throat Venturi by using DPM model is less than that by using Eulerian model, and the differential pressures in the divergent section by using the two models are analogous. Afterwards, the OR prediction correlations based on the differential pressure ratio method were proposed, and then compared and validated by the industrial field tests. The root mean square errors (RMSE) and the average relative errors predicted by Eulerian model were 4.24% and 3.78%, 5.69% and 5.01% by using DPM model, respectively. In conclusion, Eulerian model is more suitable for wet gas flow prediction. And some advice on the improvement of the multiphase flow simulation is provided to get a more preferable performance in wet gas flow prediction. © 2014 Elsevier Ltd. Source

Discover hidden collaborations