دانلود رایگان مقاله لاتین انترپولاسیون ترمودینامیکی از سایت الزویر
عنوان فارسی مقاله:
انترپولاسیون ترمودینامیکی برای شبیه سازی جریان دو فاز مخلوط غیر ایده آل
عنوان انگلیسی مقاله:
Thermodynamic interpolation for the simulation of two-phase flow of non-ideal mixtures
سال انتشار : 2016
برای دانلود رایگان مقاله انترپولاسیون ترمودینامیکی اینجا کلیک نمایید.
مقدمه انگلیسی مقاله:
1. Introduction
The modelling of compressible two-phase or flashing flows is commonplace in a wide range of areas in engineering. For example, cavitation in automotive fuel injection systems (Martynov, 2006), flash boiling of water during loss-of-coolant accidents in nuclear reactors (Blinkov et al., 1993) and liquid boiling and expansion in refrigeration systems and heat pumps (Simões-Moreira and Bullard, 2003). Whilst various approaches are available to model the dynamics of two-phase flow, the accuracy of simulations for flashing two-phase flows to a large extent depends on the accuracy of the physical properties data in use. This particularly applies to multicomponent mixtures, which are commonly found as working fluids in the above systems. In practice, complex mathematical formulas known as Equations of State (EoS), are used to provide the thermodynamic properties for both vapour and liquid phases. As a result, a prac-tical problem arises when pressure explicit EoS are coupled with flow models. In these EoS, the thermodynamic properties are predicted as a function of pressure, temperature and composition while phase equilibria, at a given system pressure and temperature (P-T), is determined using a variety of isothermal ‘flash’ algorithms (Michelsen, 1982a,b). This formulation contrasts with the fluid-dynamics models, where the conservation laws governing are naturally posed in terms of density and internal energy (-e). To overcome this problem, one possible solution is to use the so-called isochoric-isoenergetic flash (Castier, 2009; Michelsen, 1999). However, existing isochoric-isoenergetic flash algorithms are neither robust nor computationally efficient in the context of flow simulation because they either rely on an internal iterative loop over the P-T variables (Michelsen, 1999; Saha and Carroll, 1997) or on the direct minimisation of total entropy (Castier, 2009; Munkejord and Hammer, 2015). The problem of computational inefficiency is exacerbated by the complexity of modern EoS. In the oil and gas industry, fluid flow simulators almost exclusively use cubic EoS (such as SoaveRedlich-Kwong, Peng-Robinson, etc., Zaydullin et al., 2014). Cubic EoS can be solved relatively at low computational cost; however, the higher order EoS developed in the last two decades provide improved accuracy in physical property estimations, though this is at slightly higher computational cost. For example, for the case of CO2 and its mixtures, high accuracy is provided by the Statistical Associating Fluid Theory (SAFT) EoS (Diamantonis and Economou, 2011) or by the “reference” EoS (Span and Wagner, 1996) and both EoS have a large number of relatively complex terms. As a result, application of these EoS to flow simulations not only increases the computational cost, but also the susceptibility to numerical instabilities in the underlying isothermal flash algorithms. This is a particular problem in the simulation of flows associated with Carbon Capture and Storage (CCS), where the CO2 stream may contain a number of impurities and vary in composition (Porter et al., 2015). Indeed, it is well established that the presence of these impurities has important impacts on many aspects where the modelling of two-phase flow is relevant, including ductile fracture (Mahgerefteh et al., 2012a) and the release rate in the case of loss of containment (Brown et al., 2013).
برای دانلود رایگان مقاله انترپولاسیون ترمودینامیکی اینجا کلیک نمایید.
کلمات کلیدی:
22222
