46927-G9
Characterizing Multiphase Transport and Flow Dynamics in Complex Natural Gas Network Topologies
ANNUAL NARRATIVE PROGRESS REPORT: 9/1/2007 – 8/31/2008
A. TABLE OF CONTENT:
A. TABLE OF CONTENT ……………………………………………. 1
B. EXECUTIVE SUMMARY …………………………………………. 1
C. PROGRESS TO DATE ……………………………………………. 1
D. WORK IN PROGRESS ……………………………………………. 2
D. IMPACT ……………………………………………………………...... 3
B. EXECUTIVE SUMMARY:
In this work, a comprehensive research study is carried out for the analysis of the underlying physical factors that control multiphase transport and flow dynamics in complex natural gas network topologies. The behavior associated with highly interlinked, closed-loop natural gas networks is greatly scrutinized in order to denude all fundamental relationships, critical links, and major parameters controlling performance in order to achieve the characterization of the most relevant transport features. The first year of study has been dedicated to the formulation of proper governing equations coupled with thermodynamics, implementation of FVM schemes, and application to single-phase gas networks. The second year of the study will be dedicated to the extension of the model to two-phase flow conditions and the analysis of the associated uneven split of phases at the T-junctions.
C. PROGRESS TO DATE:
In this first year of work, the study of the proper formulation of governing equations was undertaken. The necessity of simultaneous solution of all the cells (control volumes) in a pipeline was emphasized from the stand point of network analysis and the foundation for multiphase flow studies have been laid out. The importance of the finite volume method (FVM) in terms of conservative property and associated convenience of utilizing lesser number of cells have been shown for both single phase and two-phase analysis. Our studies showed that closed networks, involving several loops, could be modeled using the FVM model where junctions are replaced with tee structures. Utilization of the staggered grid approach, although poses some inconvenience in terms of boundary conditions, worked very well for the modeling of single phase flow at T-junctions in single phase conditions, allowing the FVM model to be easily extended for network solution without any special treatment. During this first year, an integrated pipe approach that utilizes the Finite Volume Method on a staggered mesh was fully formulated as its advantages over other modeling approaches clearly defined. The potential advantages of using integrated pipe methods were evaluated for the case of natural gas multiphase flow in a single pipe. In all scenarios studied, the proposed integrated method outperformed marching algorithms in terms of true conservation of mass and energy. The undesirable loss of mass and energy associated with marching algorithms can also trigger the more serious problem of unrealistic pressure, temperature, and holdup profiles that can result from such predictions. It has been demonstrated that marching algorithms can fail to capture the physics of the problem and the true nature of the information conveyed by the governing equations.
D. WORK IN PROGRESS:
The next stage of the study is going to focus on the multiphase analysis in networks. Nonetheless, there are several issues to be addressed before and during the implementation of two-phase model. To begin with, proper set up and validation of the tee structure for different angles of connection, different side line (branch) diameters as well as the effect of energy loss effects, which is needed to be accounted for both compressible and incompressible single phase flows. Later, two-phase flow related problems are to be addressed. Flow pattern transitions and phase appearance-disappearance are to be studied in a straight pipe. This part of the study shall provide more insight on the possible complications of using larger cells to capture two-phase flow since one or more flow pattern transitions might be taking place within a single large cell. A mass transfer model must be proposed that can reliable capture condensate/gas interaction, for which different source/sink representations will be tested in order to best capture the nature of the phenomenon. Following the completion of robust two-phase flow model for single pipes, uneven split of phases at a T-junction must be studied. An important point to consider here is the change in overall mixture composition downstream of the junction due to the uneven split. Therefore, model has to be modified in order to accommodate this composition change after a split at a tee. Available published data will be utilized to obtain new insights into the complex phenomena of liquid branching utilizing the proposed approach, which will be based on the fundamental conservation equations.
E. IMPACT:
The ACS Type G Grant has provided the PI with critical resources at this early stage of his career to enable him to make steady and significant progress in his scholarship of research and enabled each member of his research group to maintain significant progress while working towards their academic goals.
