Litcius/Paper detail

Oil Production Optimization Using Q-Learning Approach

Mazyar Zahedi-Seresht, Bahram Sadeghi Bigham, Shahrzad Khosravi, Hoda Nikpour

2024Processes16 citationsDOIOpen Access PDF

Abstract

This paper presents an approach for optimizing the oil recovery factor by determining initial oil production rates. The proposed method utilizes the Q-learning method and the reservoir simulator (Eclipse 100) to achieve the desired objective. The system identifies the most efficient initial oil production rates by conducting a sufficient number of iterations for various initial oil production rates. To validate the effectiveness of the proposed approach, a case study is conducted using a numerical reservoir model (SPE9) with simplified configurations of two producer wells and one injection well. The simulation results highlight the capabilities of the Q-learning method in assisting reservoir engineers by enhancing the recommended initial rates.

Topics & Concepts

Oil productionEclipseProduction (economics)Petroleum engineeringProduction rateComputer scienceOil wellProduction system (computer science)Mathematical optimizationProcess engineeringEngineeringMathematicsAstronomyMacroeconomicsEconomicsPhysicsReservoir Engineering and Simulation MethodsOil and Gas Production TechniquesHydraulic Fracturing and Reservoir Analysis
Oil Production Optimization Using Q-Learning Approach | Litcius