Loading...

Development of an Object Oriented Framework for Compositional Simulation with Dynamic Component Retrieval

Assareh, Mehdi | 2013

447 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 44545 (06)
  4. University: Sharif University of Technology
  5. Department: Chemical and Petroleum Engineering
  6. Advisor(s): Ghotbi, Cyrus; Pishvaie, Mahmoud Reza
  7. Abstract:
  8. Phase behavior calculations and compositional simulation for the hydrocarbon systems containing many components, are time consuming and impractical. Therefore the number of the components those describe the fluid description must be kept to a minimum using grouping techniques. In this research, it is tried to apply a new set of grouping techniques in compositional simulation of oil and gas reservoir. In these methods, the components are automatically placed in the pseudo-components, and their critical and thermo-physical properties are calculated simultaneously or after grouping process according to different mixing rules. These set of techniques areworking based on the distances between components and pseudo-components (in different properties including critical temperature, critical pressure, acentric factor, molecular weight ...) andhave a good insight to critical and thermo-physical properties and reduces loss of information in grouping process. In this respect, different types of clustering techniques are investigated and their advantages and disadvantages are studied to find the optimized solution. Then, these algorithms are implemented in PVT and phase behavior calculation for the reservoir fluids to analyze their performance under physical feedback. To achieve this goal, a large number of detailed fluid description and chromatographic results for petroleum reservoir fluid are used to evaluate this new class of lumping techniques. To examine the quality of the prepared fluid description for fluid flow modeling, the compositional simulations are performed for the lumped and detailed fluid description. The next goal for this thesis is to retrieve the detailed fluid description based on the equilibrium calculation on the grouped system, which is called component retrieval or delumping or inverse lumping. To do so, initially, the corresponding delumping techniquesare developed and implemented to retrieve the detailed components information from pseudo-components in equilibrium calculations. Then the component retrieval is studied in compositional simulation. For this purpose, a compositional simulation research code is preparedfor three phase three dimension fluid flow modeling using a finite volume approach. The program codes developmentis under an object oriented framework to ensure its performance, flexibility and extensibility.Finally a new analytical framework for equilibrium calculation reduction is presented in this work. This reduction framework is successfully implemented for different types of reservoir fluid in different range of thermodynamic condition in the two phase region. As demonstrated in different parts of this research, the new presented grouping techniques have suitable exactness in a way that the fluid phase behavior for the lumped systems is very closed to detailed system ones. This means that reduction in fluid description components number keep the equation of states parameters flexibility for future calculations.In addition, comparison of the detailed system equilibrium ratios and the equilibrium ratios retrieved by proposed delumping technique, showed acceptable exactness. This comparison was made both in component retrieval equilibrium calculations and dynamic components retrieval in compositional simulation. The proposed techniques for grouping and components retrieval in cubic equations of states have enough capabilities to be applied in non-cubic equations of state. Therefore, developed techniques are also applied to a candidate non-cubic equation of state derived from statistical associated fluid theory
  9. Keywords:
  10. Compositional Simulation ; State Equation ; Components Lumping ; Component Retrieval ; Object Oriented Simulation ; Gas Reservoirs ; Oil Tanks

 Digital Object List

 Bookmark

No TOC