The Ideal Range for Reducing Oil Viscosity Through Hot Water Injection in a Heavy Oil Field

The most effective methods for enhancing the extraction of hydrocarbons, especially in heavy oil reservoirs, involve the application of thermal techniques. Thermal enhanced oil recovery (EOR) involves the injection of steam or hot fluids into subsurface reservoirs to alter the physical properties, such as fluid viscosity and effective mobility. While various mathematical models have been developed to estimate temperature changes during these processes, they often assume constant fluid velocity throughout the reservoir, leading to significant inaccuracies in the predictions. This study aims to introduce a novel approach to accurately measure temperature distribution, heat propagation over time, and distance from the wellbore in thermal EOR operations. The primary objective of this research is to develop a robust model capable of precisely predicting temperature distributions within porous rock formations during thermal EOR procedures. Artificial intelligence (AI) techniques were employed to compute temperature profiles, reducing the complexity and unpredictability associated with numerical methods. The model’s efficiency was enhanced by considering factors like formation permeability, injection time, and distance from the wellbore. To determine temperature distribution, neural networks, fuzzy logic systems, and generalized intelligent networks were utilized. The intelligent networks were fine-tuned by adjusting various model parameters, and the model’s effectiveness was assessed using metrics such as the average absolute error and correlation coefficient.