Ajuste fino de correlaciones PVT para algunos crudos colombianos
No Thumbnail Available
Date
Advisors
Evaluators
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad Industrial de Santander
Abstract
Description
La estimación de las propiedades del fluido de un yacimiento y su variación en función de la Temperatura y la Presión representa un elemento clave para construir un modelo de simulación de flujo de producción acertado. El método de modelamiento más riguroso actualmente aplicado utiliza las Ecuaciones de Estado (EOS), partiendo de una descripción detallada de la composición del fluido y la asignación de ciertas propiedades para cada uno de sus componentes. Una aproximación más generalizada considera el fluido divido únicamente en tres fases: agua, crudo y gas, conocido como Black-Oil, donde generalmente solo se tiene en cuenta la solubilidad del gas en las dos fases líquidas, mientras el agua y el aceite se consideran completamente inmiscibles. El desarrollo de cualquiera de estos dos tipos de modelo se basa en el ajuste a pruebas PVT realizadas en laboratorio.
Los esfuerzos de modelado del fluido están generalmente relacionados con su propósito y las características del mecanismo de producción, de forma que frecuentemente la simulación de fenómenos a nivel de pozo y superficie (no así a nivel de yacimiento), suelen simplificarse por el uso de correlaciones empíricas basadas en el modelo Black-Oil, las cuales son de amplio uso y cuentan con numerosas propuestas de diferentes autores. Estas correlaciones pueden fallar de forma crítica cuando son extrapoladas a condiciones de presión y temperatura diferentes a las de las pruebas experimentales. En este trabajo se presentan los resultados de una metodología para el ajuste fino mediante optimización de diferentes correlaciones para la predicción del comportamiento PVT de los fluidos de cuatro campos diferentes de Colombia, con características de crudo pesado (Campos Castilla y Chichimene) y otros más livianos (Campos Apiay y Yariguí,), cuyas gravedades API se encuentran entre 7-30 °. A la información experimental de prueba PVT se añaden además restricciones de comportamiento esperado y propiedades del fluido a condiciones de superficie, obteniendo expresiones que logran representar congruentemente fluidos de yacimiento en condiciones típicas operacional de un pozo productor, y que sirven no solo para los crudos ajustados sino también para crudos con propiedades similares, como por ejemplo del mismo campo
The accurate estimation of reservoir fluid properties as a function of temperature and pressure depicts a key step in the procedure of developing a correct well flow model. The strictest model currently applied uses the Equations of State (EOS), by a detailed description of the fluid composition and assigning certain properties for each component. One generalized approach so called Black-Oil model considers the reservoir fluid as mixture of three phases: water, oil and gas, which generally regards only the dissolution of gas in water and oil, but the liquids are immiscible. Both approaches use adjusting of experimental PVT tests to mathematical expressions. Generally, model developing effort is related to the purpose of the fluid model and production mechanism, nevertheless in bottom and surface well level cases (not for reservoir level), the estimation is simplified using empirical correlations based on Black-Oil model which have been widely used and several authors develop them. These correlations lack in extrapolated conditions. In this work, the results of fine tuning methodology through different correlations optimization to PVT properties of four Colombian crude oils (Castilla and Chichimene as heavy and extra-heavy oils, and other more lighter crudes such as Apiay and Yariguí), whose API gravities are between 7 to 30°. Experimental PVT data test and other conditions such as well surface data and expected performance are used to obtain expressions which achieve good representation of fluid properties in typical well operational conditions. These expressions are useful for other fluids with comparable properties than the fluid used for adjusting, for instance well fluids of the same field.
The accurate estimation of reservoir fluid properties as a function of temperature and pressure depicts a key step in the procedure of developing a correct well flow model. The strictest model currently applied uses the Equations of State (EOS), by a detailed description of the fluid composition and assigning certain properties for each component. One generalized approach so called Black-Oil model considers the reservoir fluid as mixture of three phases: water, oil and gas, which generally regards only the dissolution of gas in water and oil, but the liquids are immiscible. Both approaches use adjusting of experimental PVT tests to mathematical expressions. Generally, model developing effort is related to the purpose of the fluid model and production mechanism, nevertheless in bottom and surface well level cases (not for reservoir level), the estimation is simplified using empirical correlations based on Black-Oil model which have been widely used and several authors develop them. These correlations lack in extrapolated conditions. In this work, the results of fine tuning methodology through different correlations optimization to PVT properties of four Colombian crude oils (Castilla and Chichimene as heavy and extra-heavy oils, and other more lighter crudes such as Apiay and Yariguí), whose API gravities are between 7 to 30°. Experimental PVT data test and other conditions such as well surface data and expected performance are used to obtain expressions which achieve good representation of fluid properties in typical well operational conditions. These expressions are useful for other fluids with comparable properties than the fluid used for adjusting, for instance well fluids of the same field.
Keywords
Correlation Optimization, Viscosity, Density, Volumetric Factor, Optimización Correlaciones, Viscosidad, Densidad, Factor Volumétrico