Assessment of the Accuracy and Dynamic Simulation Capabilities of Liquid-Vapour Two-Phase Flow Separated and Mixture Models

Abstract

Liquid-vapour two-phase flows through pipes, ducts and channels are encountered in almost all chemical process plants and industrial applications. Efficient and accurate numerical methods must be applied to two-phase flow mathematical models that are just as accurate if such process facilities and equipment are to be properly designed and simulated. This problem has been currently tackled by means of one-dimensional simulations based predominantly on Two-Fluid (Separated) and Drift-Flux (Mixture) Models, although simpler methods which are based on the assumption of homogeneous flow are also seen to persist. Given this context, the goals of the present study are twofold. In order to improve and consolidate the model selection process, Differential- Algebraic Equations (DAE) numerical techniques were combined with the steady-state mass and momentum conservation equations of the Two-Fluid, Drift-Flux and Homogeneous models, and also with numerous constitutive equations for relevant quantities such as drag coefficients and specific interfacial area, in the reproduction of experimental void fraction and pressure drop data comprising the most important flow regimes as well as the entire range of inclination angles between upward and downward orientations. In this assessment of experimental data reproducibility, most accurate predictions were obtained with the Drift-Flux Model, thus making that approach even more attractive given its simpler formulation as compared to the Two-Fluid Model. Subsequently, extension of the DAE numerical approach to dynamic two-phase flow simulations is demonstrated. The proposed technique was applied to a test case of a multicomponent hydrocarbon transient two-phase flow including mass and heat transfer. This scenario was successfully investigated by means of dynamic simulations based on the Two-Fluid and Drift-Flux models, both including rigorous calculations of flash and thermophysical and transport properties as well as truncation error estimates for stepsize selection. Such results are expected to encourage the application of DAE numerical methods to both steady and unsteady-state one-dimensional two-phase flow calculations.