An Investigation into Sub-Critical Choke Flow Performance in High Rate Gas Condensate Wells

Abstract

There have been some correlations in the literature to predict the gas and liquid flow rate through wellhead chokes under subcritical flow conditions. The majority of these empirical correlations have been developed based on limited production data sets that were collected from a small number of fields. Therefore, these correlations are valid within the parameter variation ranges of those fields. If such correlations are used elsewhere for the prediction of the subcritical choke flow performance of the other fields, significant errors will occur. Additionally, there are only a few empirical correlations for sub-critical choke flow performance in high rate gas condensate wells. These led the authors to develop a new empirical correlation based on a wider production data set from different gas condensate fields in the world; 234 production data points were collected from a large number of production wells in twenty different gas condensate fields with diverse reservoir conditions and different production histories. A non-linear regression analysis method was applied to their production. The new correlation was validated with a new set of data points from some other production wells to confirm the accuracy of the established correlation. The results show that the new correlation had minimal errors and predicted the gas flow rate more accurately than the other three existing models over a wider range of parameter variation ranges.

Robust hybrid machine learning algorithms for gas flow rates prediction through wellhead chokes in gas condensate fields

“…They adapted the Gilbert equation in three modified forms by changing the pressure parameters (replacing the upstream pressure with the pressure drop across the choke) for the wells' data in gas condensate reservoirs. [30]. Seidi and Sayahi (2015), by adapting Gilbert's basic equation using the genetic algorithm and nonlinear regression methods and applying them to 67 datasets gathered from different gas condensate fields, proposed an optimized model for estimating the condensate gas flow rate [31].…”

Modelling oil and gas flow rate through chokes: A critical review of extant models

“…While most studies considered three input parameters (wellhead pressure, choke size and GOR), others included parameters such as water cut, BS&W, API gravity and gas liquid ratio. The widest range of wellhead temperature was in the study by Nasriani et al (2019) wherein the temperature ranges from 191 to 463 • F, while the least range was 60-120 • F as given by Abdul-Majeed (1988). For all wellhead pressure ranges, the highest [1400-12000 psia] was that given by , while the least [60-350 psia] was by Ganat and Hrairi (2018).…”

“…Ganat and Hrairi (2018) Non-linear regression The developed model is quick, reliable, and can be adapted to any ESP oil well as well as artificially onshore and offshore flowing wells but can only be applied to critical flow conditions. Fuladgar and Vatani (2019) Multi variable linear regression An evaluation of the new correlation indicated that it could significantly improve accuracy of flowrate predictions in contrast to previous prominent correlations Nasriani et al (2019) Non-linear regression This model works best when applied within the range of the following production parameters: LGR of 0.7-178.8 bbl/MMscf, a choke size: 24/64 to 192/64, gas flow rate of 5.4-113.3 MMscfD. intelligence, a seamless merger of two or more machine learning techniques with the aim of complementing each other is called hybrid computational intelligence or hybrid intelligent systems (Anifowose, 2011).…”

“…For all wellhead pressure ranges, the highest [1400-12000 psia] was that given by , while the least [60-350 psia] was by Ganat and Hrairi (2018). For choke size range, the highest [37-192(1/64 inch)] was by Nasriani et al (2019),…”