Optimization of WAG in real geological field using rigorous soft computing techniques and nature-inspired algorithms

To meet the ever-increasing global energy demands, it is more necessary than ever to ensure increments in recovery factors (RF) associated with oil reservoirs. Owing this challenge, enhanced (EOR) techniques are increasingly gaining significance as robust strategies for producing volumes from mature Water alternating gas (WAG) injection an EOR method intended at improving microscopic and macroscopic displacement efficiencies. handle implement successfully technique, of vital importance optimize its operating parameters. This study targeted implementing proxy paradigms investigating suitable design parameters a WAG project applied real field data “Gullfaks” North Sea. The models aimed reducing significantly rum-time related commercial simulators without scarifying accuracy. end, machine learning (ML) approaches, including multi-layer perceptron (MLP) radial basis function neural network (RBFNN) were implemented estimating needed formulated optimization problem. improve reliability these ML methods, they evolved using algorithms, namely Levenberg–Marquardt (LM) MLP, ant colony (ACO) grey wolf (GWO) RBFNN. performance analysis revealed that MLP-LMA has better prediction ability other two paradigms. In context, highest average absolute relative deviation noticed per runs by was lower 3.60%. Besides, best-implemented coupled ACO GWO resolving studied findings suggested proxies cheap, accurate, practical emulating numerical reservoir model. addition, results demonstrated effectiveness optimizing process used study.