CO 2 has been a significant part of the advancement and development of hydraulic fracturing. This technology was proven in the 1970s as an energy source to help recovery of well stimulation fluids and is predominantly used to stimulate tight sandstone reservoirs because it is both clean and effective. The first application of CO 2 in Saudi Arabia is discussed, highlighting its success in terms of 50% increased productivity, quick cleanup, and reduced water volume.Classically, conventional hydraulic fracturing (acid or proppant) has been used to maximize long-term production and minimize near wellbore (NWB) damage. This paper discusses how CO 2 fracturing can offer more advantages by reducing or eliminating the use of the underground water. It also provides a gas drive solution for effective cleanup. CO 2 can be used to significantly reduce interfacial tension and resultant capillary forces, thus helping remove fracturing fluid, connate water, and emulsion blocks, which are among the primary damage mechanisms when fracturing gas wells. CO 2 has been a significant part of the advancement and development of hydraulic fracturing since the 1970s, but mostly just in the US during those early days. This paper discusses in detail the first application in Saudi Arabia using CO 2 to foam a proppant fracturing treatment. The implementation of this technology was successful. This was demonstrated by the resultant 50% productivity increase in the study well. Using this method allowed a very fast and effective cleanup compared to other wells fractured using conventional crosslinked fluid methods. Additionally, using this method provided the added benefit of reducing water volume, which is a critical factor in this area because of limited water availability.The first CO 2 application in Saudi Arabia is discussed, highlighting its success in a resultant 50% productivity increase. The method allowed fast and effective cleanup compared to wells fractured using conventional crosslinked fluids. Additional benefits included reduced water volume, which was critical in this area because of limited resources.

In this study, field data from Saudi Arabian gas condensate reservoirs were used to develop Artificial Neural Network (ANN) model to estimate gas rate from wells having Condensate to Gas Ratio (CGR) ranging from 10 to 400 STB/MMSCF. This model can predict gas rates for both critical and subcritical flow regimes. This paper will discuss the advantages of this new model as well as the limitations of utilizing this model in predicting the gas rate. Most of the choke correlations estimate gas rates with high accuracy in dry gas environment or in reservoirs with low CGR. There are some correlations built to estimate the rates for multiphase flow; however, there is no unique correlation that works for wide range of CGR values with acceptable accuracy. Also, there are two sets of correlations were built for critical flow regime and for subcritical flow. The ANN model was developed based on analyzing more than 6,000 data points collected from separator tests. These data were used as follows: 70% for training, 15% for validation, and 15% for testing. The data covers wide range of flow rates, choke sizes, and CGR values. The models were built as function of choke sizes, upstream & downstream pressures, and CGR. Graphical and statistical tools have been used to evaluate and validate the ANN models. In conclusion, the new ANN model applies novel techniques and wide range of field data and predicts the gas rate for high CGR wells with an average absolute error of 10.2 %. These models will be used to forecast and evaluate performances of gas reservoirs and will minimize utilizing the costly separator tests normally required high CGR wells.