دانلود رایگان مقاله انترپولاسیون ترمودینامیکی برای شبیه سازی جریان دو فاز مخلوط غیر ایده آل

Abstract

This paper describes the development and application of a technique for the rapid interpolation of thermodynamic properties of mixtures for the purposes of simulating two-phase flow. The technique is based on adaptive inverse interpolation and can be applied to any Equation of State and multicomponent mixture. Following analysis of its accuracy, the method is coupled with a two-phase flow model, based on the homogeneous equilibrium mixture assumption, and applied to the simulation of flows of carbon dioxide (CO2) rich mixtures. This coupled flow model is used to simulate the experimental decompression of binary and quinternary mixtures. It is found that the predictions are in good agreement with the experimental data and that the interpolation approach provides a flexible, robust means of obtaining thermodynamic properties for use in flow models.

4. Conclusions

This paper presents the development and application of a robust interpolation technique for the prediction ofthermodynamic properties and phase equilibria of complex mixtures. The accuracy and computational burden of computing these physical properties greatly affects the overall accuracy and computational cost of multiphase multicomponent simulations. Thus, the adaption of this technique has a tremendous impact on our ability to perform sophisticated computational fluids dynamics (CFD) simulations at reasonable cost without significant loss of accuracy. Furthermore, in this work, the higher order PC-SAFT EoS was used for the accurate calculation of physical properties, to the author’s knowledge it is the first attempt to use PC-SAFT for dynamic, compressible, multiphase fluid flow simulations. The assessment of the technique’s ability to reproduce the results of the EoS showed, for the most part, an error no greater than 0.5% compared to the actual EoS predictions. Large errors were observed only for the liquid phase at low temperatures, where the physical model represented by the EoS is itself not applicable, as solid formation not predicted by the EoS is expected. The extension of the current interpolation technique to a thermophysical model where the solid phase is accounted for is part of ongoing work. Following this, the method was coupled with a fluid model and was used for the simulation of CO2 rich mixtures, which is of particular interest in the development of CCS technology. Analysis of several hypothetical shock tube tests, as well as the comparison of the predictions against experimental decompression data, showed that the interpolation method produced robust and highly reliable results for simple and complex mixtures. Interestingly, comparison between model predictions and experimental decompression results showed that the implementation of the interpolation technique produced a reasonable prediction of the initial depressurisation period. On-going work by the authors focuses on the development of appropriate models for the heat transfer and frictional effects to improve the accuracy of the predictions beyond this period.