Screening and Selection of Artificial Lift Systems Using Artificial Intelligence Technique for a given Well

Abstract

In spite of numerous studies over the past few decades to comprehend tight formations and create cutting-edge technologies to optimise the application of artificial lifts, there are still varying opinions on the best strategy, Artificial Lift (A.L.) type, and ideal circumstances for installing artificial lifts throughout the life of a well. This research thoroughly analyses the use of artificial lift systems with a particular focus on multi-fractured formations. The review focuses on two recent unconventional horizontal wells called Osprey and Hawk. The goal is to screen and design an optimal and cost-effective Artificial Lift Technology that best fits the Wells in this study. An Artificial Intelligence Screening model was built using Python programming language and supported by Random Forest Algorithm. Productivity analysis and the design specification of the selected lift system after the screening was done using PROSPER Simulation Software. Gas Lift was selected for Osprey well with a productivity Increment of 67%. With regards to the Hawk field, Gas lift was also selected with a productivity increment of 47%. The implementation of any of the optimization strategies for the gas lift will enhance profitability, reduce operational costs, and extend the life of the wells. However, it is recommended to run additional simulations for additional lift methods and use them with the decision matrix, and also keep adding more criteria to the decision matrix that have an impact on screening and selection.