Enhanced Oil Recovery Efficiency Using AI-Optimized Polymer Flooding in Mature Reservoirs
Info: Enhanced Oil Recovery Efficiency Using AI-Optimized Polymer Flooding in Mature Reservoirs Topics I phdassistance.com
Published: 12th August in Enhanced Oil Recovery Efficiency Using AI-Optimized Polymer Flooding in Mature Reservoirs Topics I phdassistance.com
Share this:
Related Services
Our academic writing and marking services can help you
Introduction
The growing necessity to recover oil efficiently from mature reservoirs has led to the incorporation of modern technology with enhanced oil recovery processes. With conventional flooding techniques becoming inefficient because of decreasing pressures, high water cuts, and oil left behind in non-uniform rock formations, there is a rising need for intelligent methods that can help in recovering oil efficiently. Modern breakthroughs in the field of artificial intelligence, machine learning, reservoir simulation, and prediction analysis have opened up new avenues in this regard. Several areas require attention, such as reservoir performance prediction, managing geological uncertainty, controlling injection, choice of polymer formulation, and field simulation. From current literature, there is a need for increased use of artificial intelligence in conjunction with polymer-based EOR, especially with AI-Enhanced Oil Recovery.
Proposed PhD Topic 1: Explainable Machine Learning for Identifying Reservoir and Fluid Mechanisms Governing Enhanced Oil Recovery in Mature Fields
Background Context:
The application of machine learning techniques is rapidly growing to analyse and forecast reservoir behaviour and optimise oil recovery. Nourizadeh et al. (2026) used Feedforward Neural Networks and Elman Recurrent Neural Networks to forecast the oil recovery and water cut behaviour during the polymer flooding process. This example proved the ability of using AI techniques to model the non-linear relationship between reservoir properties and operating parameters.
However, the high level of predictive accuracy cannot be the explanation for the success of certain reservoir and fluid conditions in terms of oil recovery. There are several variables such as permeability, polymer concentration, mobility ratio, viscosity and injection conditions that may have a complicated interrelation. This gives a chance to go further from prediction and apply techniques of explainable machine learning where the influence and interaction of each variable could be understood from the point of view of reservoir engineering. The research could be expanded on the topic of Polymer Flooding Optimization.
PhD-Level Verification:
Current literature has already proven the capability of machine learning algorithms in predicting oil recovery and water cut behaviour during the polymer flood process. Yet, even with prediction accuracy, little is known about how the underlying physics behind those results is related. There is a need for research regarding the factors influencing oil recovery and how these interact within the reservoir environment. A PhD research project may be undertaken involving the use of explainable machine learning and principles of reservoir engineering to determine the significant variables.
Research Questions:
Contributions at the PhD-Level:
Suggested Readings:
Nourizadeh, M., Khosravi, R., Simjoo, M., & Chahardowli, M. (2026). A hybrid AI-genetic algorithm framework for the optimisation of polymer flooding strategies: A numerical simulation-based approach.
Proposed PhD Topic 2: Robust Decision-Making Under Geological Uncertainty for Chemical Flooding Performance in Heterogeneous Mature Oil Reservoirs
Background Context:
Heterogeneity of the reservoir is one of the biggest obstacles to EOR techniques since changes in permeability and porosity can affect flow behaviour significantly. In their study, Zhong et al. (2018) managed to illustrate the difficulties that arise while attempting to model polymer flooding in large heterogeneous reservoirs and how distributed high-performance computing can help overcome such issues. However, even high-resolution models of the reservoir are unable to remove all geological uncertainties. In particular, the actual values of permeability, residual oil saturation, and reservoir connectivity can be different from those assumed in the model. This means that the same flooding policy will lead to vastly different results if applied in different geological realisations. In this context, there is an interesting research topic related to Enhanced Oil Recovery Optimization in a risk-aware way, when various geological realisations are considered and compared in order to choose the best strategy.
PhD-Level Verification:
The current research on large-scale polymer flooding clearly shows the significance of using sophisticated computation techniques for reservoir simulation, especially for heterogeneous reservoirs. Yet, the accurate simulation of reservoirs still does not eliminate the uncertainties related to permeability distributions, connectivity, the saturation level of residual oil, and many other geological properties of the reservoir. Not much work has been done in the direction of applying multiple geological realisations for finding the recovery strategies which are robust against uncertainty.
Research Questions:
PhD-Level Contributions:
Suggested Readings:
Zhong, H., Liu, H., Cui, T., Shen, L., Yang, B., He, R., & Chen, Z. (2018). Numerical Simulations of Polymer Flooding Process in Porous Media on Distributed-memory Parallel Computers.
Proposed Dissertation topic 3: Reinforcement Learning for Adaptive Injection Control and Water-Cut Management During Late-Life Production from Mature Oil Reservoirs
Background Context:
Reservoir management becomes more complicated as water production increases and the amount of easy-to-produce oil decreases. What is successful in terms of injection at the start of a flooding program may not be relevant later on, as pressure, water cut, fluids, and oil saturation change. This calls for more flexible strategies for managing the reservoir.
Li et al. (2025) studied discontinuous chemical flooding using various polymer and gel types and found that the effects of slug size and chemical sequence switching on displacement can be affected by the conditions of heterogeneity of the reservoir. The results obtained from the experiment suggest that chemical injection timing can be a crucial parameter.
However, traditional flooding plans are usually developed beforehand. They may not always adapt themselves to changing production behaviour. This presents an opportunity to use reinforcement learning where the intelligent system can learn about changes in production and learn how injection conditions can vary with time during later stages of reservoir production. The study can therefore consider AI Reservoir Optimization as an adaptive control problem and not a single optimisation problem.
PhD Level Verification:
The existing studies demonstrate that the dimensions and switching sequences of polymer and gel slugs affect the effectiveness of displacement processes in heterogeneous reservoirs. At the same time, all these parameters are chosen according to pre-established schemes regardless of the changes in production characteristics. The research gap thus is related to the lack of studies devoted to the creation of intelligent methods for controlling the process on the basis of analysis of production data.
Research Questions:
PhD-Level Contributions:
Suggested Readings:
Li, X., Zhang, J., Zhang, Y., et al. (2025). Experimental Study on the Application of Polymer Agents in Offshore Oil Fields: Optimisation Design for Enhanced Oil Recovery.
Proposed Dissertation Topic 4: Data-Driven Design of Reservoir-Specific Viscosity-Enhancing Agents for Improving Oil Displacement Efficiency in Mature Reservoirs
Background Context:
Polymer flooding efficiency does not only depend on injection conditions but also on the physical and chemical characteristics of the polymer used. Characteristics such as viscosity, concentration, adsorption, retention, temperature, salinity, and interactions with the porous media may affect the mobility control ability and oil displacement.
In their study, Yang et al. (2023) examined a new type of phase transition polymer in mature waterflooding reservoirs. According to the results of their experiments and microscale analysis, the structure and viscosity behaviour of polymers can enhance the displacement efficiency and sweep efficiency as compared to conventional polymeric materials.
However, despite the ability to show the possibility of using modern polymers, there is still difficulty in choosing the best type of polymer for a certain reservoir. Polymers that work efficiently at one temperature, permeability or salinity will definitely not have the same efficiency at other places. This provides room for further research based on data that will link polymers’ properties with reservoir properties and thus formulate more suitable polymers for particular mature reservoirs. This method can be used as the basis for enhancing PFE in future.
PhD-Level Verification:
Experiments involving mature phase transition polymers have demonstrated that structural variations in the polymer, together with viscosity and age-related properties, may enhance the displacement capability. Nevertheless, formulation selection still largely relies on laboratory studies and reservoir testing. There is a lack of research into data-based correlations of polymer and reservoir parameters. An area for future PhD investigation would be to build predictive models of compatibility between the polymers and the reservoir.
Research Questions:
Contributions at the PhD-Level:
Suggested Readings:
Yang, Y., Cao, X., Ji, Y., & Zou, R. (2023). Mechanistic Insights into a Novel Controllable Phase-Transition Polymer for Enhanced Oil Recovery in Mature Waterflooding Reservoirs.
Proposed Dissertation Topic 5: Digital Twin-Based Intelligent Management of Full-Field Enhanced Oil Recovery in Complex and Highly Heterogeneous Mature Reservoirs
Background Context:
Managing mature fields in an integrated manner needs consistent knowledge of production behaviour, reservoir state, and distribution of oil in place. Nevertheless, high-resolution reservoir simulation becomes computationally intensive, especially where many scenarios and different conditions of operation have to be simulated.
Gomaa et al. (2025) showed that the predictive capability of machine learning techniques such as Artificial Neural Network, Random Forest, K-Nearest Neighbour, and Support Vector Machine can be used to predict total oil recovery from parameters of the reservoir and waterflooding process. Based on this work, further research at the doctoral level may result in the development of a digital twin that would use the historical data of production and reservoir simulation along with surrogate models generated through artificial intelligence techniques.
PhD-Level Verification:
According to Gomaa et al. (2025), machine learning models have been shown to make predictions on the total recovery of oil through reservoir and water flooding characteristics. These models usually generate predictions without giving an indication of how the reservoir is changing at any particular moment. A PhD research project could take advantage of this gap in the current knowledge by linking machine learning to reservoir simulation.
Research Questions:
PhD-Level Contributions:
Suggested Readings:
Gomaa, S., Soliman, A. A., Mansour, M., El Salamony, F. A., & Salem, K. G. (2025). Machine learning models for estimating the overall oil recovery of waterflooding operations in heterogeneous reservoirs.
Need assistance finalising your dissertation topic? Selecting a strong, researchable topic can be challenging — but you don’t have to do it alone.
Our research consultants can help refine your ideas, identify literature gaps, and guide you toward a topic that aligns with current academic trends and your programme requirements.
Contact us to begin one-on-one topic development and refinement with PhdAssistance.com Research Lab.
Share this:
Cite this work
PhDAssistance. (n.d.). Cybersecurity in business Dissertation Topics Retrieved January 28th, from https://phdassistance.com/topic/cybersecurity-business/
Jalolova, M., and Musawwir, M. “Cybersecurity in business Dissertation Topics for PhD Scholars.” PhDAssistance, https://phdassistance.com/topic/cybersecurity-business/ Accessed 28th January 2026.
Jalolova, M., and Musawwir, M. “Cybersecurity in business Dissertation Topics for PhD Scholars.” PhDAssistance, PhDAssistance, Web. 28th January 2026.
Jalolova, M., and Musawwir, M., n.d. Cybersecurity in business Dissertation Topics for PhD scholars. [online] Available at: https://phdassistance.com/topic/cybersecurity-business/ [Accessed 28th January 2026].
Jalolova M., Musawwir M. Cybersecurity in business Dissertation Topics for PhD scholars [Internet]. PhDAssistance; [cited 2026 28th January]. Available from: https://phdassistance.com/topic/cybersecurity-business/
Jalolova, M., and Musawwir, M. (n.d.). Cybersecurity in business Dissertation Topics for PhD scholars. Retrieved 28th January 2026, from https://phdassistance.com/topic/cybersecurity-business/
Jalolova, M., and Musawwir, M., Cybersecurity in business Dissertation Topics (n.d.) https://phdassistance.com/topic/cybersecurity-business/ accessed 28th January 2026.
Study Resources
Free resources to assist you with your university studies!