An intelligent solution to forecast pressure drop in a vertical well having multiphase flow using functional network technique

For a well production optimization and cost per barrel of oil reduction, it is very essential to have an accurate measurement of well flowing bottom-hole pressure (FBHP) available all the time during the life of the well. It is a common practice follow in the oil and gas industry to run bottom-hole pressure gauges to record FBHP. However, interfering a producing well is an expensive and time-consuming task, involved with production disruptions and safety risks. To address these concerns, numerous mechanistic and empirical models were formulated to predict FBHP. Among these a large number of models were developed under small laboratory scale conditions and are, ultimately, inaccurate when up-scaled to field conditions. This study presents an intelligent solution by the development of an empirical model to quantify FBHP for a vertical well. The new model is based on the surface production data such as tubing perforation depth, flow rate of oil, flow rate of gas, flow rate of water, API gravity of oil, tubing string internal diameter, well bottom-hole temperature, wellhead surface temperature, and wellhead pressure. The data used to train the proposed empirical model collected from the published sources, which covered practically reasonable values. The proposed model is also validated by testing against new dataset, and the results were then compared statistically with the other methods used in petroleum industry. The results show that the proposed empirical model considerably outperforms reviewed models and delivers the prediction of FBHP with high accuracy. The novelty of the proposed empirical model is that it depends only on the surface production which makes the prediction of FBHP in a real time. The proposed model is accurate and can serve as a handy tool for the production engineers to forecast the FBHP in a real time.