Announcing the 2026-2027 Texas A&M Chevron Energy Graduate Fellows

As next-generation energy systems, such as modular nuclear reactors and hybrid systems, grow in complexity, traditional computational workflows remain heavily reliant on labor-intensive expert knowledge. My research addresses this scalability barrier by developing advanced artificial intelligence methods that support the design, analysis, and operation of complex energy systems.

The work focuses primarily on two core research areas. The first area involves the development of specialized Large Language Model agents designed to automate complex modeling and simulation workflows, including multi-modal input generation, error diagnosis, simulation orchestration, and results interpretation. The second area utilizes scientific machine learning by integrating physics-informed neural networks (PINNs), uncertainty quantification (UQ), and reinforcement and continual learning to support modeling & simulation and experimental setups. This research aims to enhance operational efficiency and accelerate the transition toward a more reliable, lower-carbon energy future.

Zaid Abulawi is a Ph.D. candidate in Nuclear Engineering at Texas A&M University under the supervision of Dr. Yang Liu, and a TAMIDS Graduate Research Fellow at the Generative AI for Science and Engineering (GAISE) Lab. His research focuses on applying artificial intelligence to energy and engineering applications to support both computational workflows and experimental setups. Specifically, Zaid’s work integrates scientific machine learning, including physics-informed neural networks, uncertainty quantification, and reinforcement and continual learning algorithms, alongside the development of specialized Large Language Model (LLM) agents designed to automate and optimize modeling and simulation solvers.

This research investigates the degradation of subsurface materials, including wellbore cement and rock formations, under long-term exposure to reactive fluids, elevated temperature, and pressure. Such degradation critically impacts the integrity of systems used in geothermal energy and enhanced oil recovery, where direct monitoring and remediation are limited.

The study integrates laboratory experiments, field-exposed samples, and physics-based modeling to examine the coupled chemo-mechanical processes governing material deterioration. Cement retrieved from a CO2-enhanced oil recovery site and shale samples
subjected to controlled high-pressure, high-temperature conditions are characterized using advanced multiscale multi-physics experiments. These techniques are used to quantify microstructural and compositional changes that contribute to mechanical weakening across scales.

The results indicate that coupled chemical and mechanical processes significantly accelerate material degradation, particularly in quartz-rich formations, highlighting the importance of incorporating multiscale interactions in predictive models of long-term subsurface performance. A machine learning-assisted framework is also introduced to map spatial variations in mechanical properties, enabling efficient upscaling from microscale measurements to continuum-scale behavior.

Overall, this work advances understanding of long-term subsurface stability by linking microscale chemo-mechanical evolution to macroscale mechanical response, supporting improved prediction of well and reservoir integrity in subsurface energy systems. This research aligns with Chevron’s efforts to improve the reliability and safety of subsurface systems for lower-carbon energy technologies, including carbon capture and storage.

Shaziya Ahmed Banu is a Ph.D. Candidate in the Department of Civil and Environmental Engineering at Texas A&M University, College Station, Texas. Her research focuses on chemo-mechanical investigation of cementitious materials and geomaterials with applications in subsurface energy systems and environmental sustainability, under the supervision of Dr. Sara Abedi and Dr. Arash Noshadravan. Prior to joining Texas A&M University, she was employed as a Research Associate at American University of Sharjah (AUS), Sharjah, United Arab Emirates (UAE), under the supervision of Dr. Mohammad AlHamaydeh. She holds a Bachelor of Science in Civil Engineering and Master of Science in Civil Engineering from American University of Sharjah. In addition, she served for one year as a geotechnical engineering laboratory lecturer and supervisor at American University of Sharjah, teaching and mentoring approximately 50 students per semester. She is committed to advancing research and education in geotechnical engineering and sustainable infrastructure through interdisciplinary and application-driven investigations.

Ahmet Birkan Demir’s research focuses on generating hydrogen from petroleum reservoir systems using microwave technology as a potential lower-carbon energy pathway. Specifically, the study investigates how reservoir rocks, crude oil, and brine environments respond to microwave heating and how mineral–fluid interactions influence hydrogen production. A key motivation behind this research is the idea that existing petroleum reservoirs can serve not only as sources of hydrocarbons but also as engineered subsurface environments for hydrogen production. By concentrating on carbonate-rich reservoir systems, Ahmet Birkan Demir aims to develop a method that enhances hydrogen generation while also supporting CO₂ storage in subsurface formations. This work introduces a new perspective to the literature by integrating petroleum reservoir engineering, subsurface energy conversion, and lower-carbon hydrogen generation.

Ahmet Birkan Demir is a Ph.D. student and Graduate Research Assistant in Petroleum Engineering at Texas A&M University, working under the supervision of Dr. Berna Hascakir. His doctoral research is part of a U.S. Department of Energy-funded project focused on in-situ hydrogen production from subsurface reservoir systems.

Before joining Texas A&M, Ahmet worked for over 7 years as an R&D engineer at Turkish Petroleum (TPAO), where he contributed to research and development projects to solve production-related field problems across onshore and offshore operations.

He earned his M.Sc. in Petroleum and Natural Gas Engineering from West Virginia University, supported by a national scholarship from Turkish Petroleum for graduate study abroad. He also holds B.Sc. degrees in Chemical Engineering and Mechanical Engineering from Ataturk University, Türkiye.

Ahmet Birkan Demir’s background in petroleum engineering, field operations, laboratory research, and experimental problem-solving provides a strong foundation for his current research in lower-carbon subsurface energy. He has been a member of the Society of Petroleum Engineers since 2016 and currently serves as Vice President-Internal of the Turkish Student Association at Texas A&M University.

Touka Elsayed’s research develops experimentally grounded modeling workflows for CO2 Plume Geothermal and CO2 storage systems. Her work investigates how thermal shock affects the microstructure and geomechanical response of igneous rocks, including granite and ultramafic formations, using micro-CT imaging, sonic velocity measurements, elastic property evaluation, and mechanical testing. These laboratory observations are integrated into thermo-hydro-chemo-mechanical modeling frameworks to evaluate injectivity, thermal efficiency, reactive effects, and mechanical integrity during subsurface energy operations. By linking rock-scale thermal damage to reservoir-scale performance, this research supports safer and more efficient design of scalable lower-carbon energy systems, particularly those involving carbon utilization and novel geothermal solutions.

Touka Elsayed is a Ph.D. student in Petroleum Engineering at Texas A&M University. Her research focuses on sustainable subsurface energy systems, including CO2 Plume Geothermal systems, CO2 storage, experimental rock characterization, and thermo-hydro-chemo-mechanical modeling. During her Master’s research, she worked on reservoir-scale modeling of CO2 Plume Geothermal systems and technoeconomic analysis of carbon capture and storage applications. Her current Ph.D. work investigates thermal shock-induced damage in igneous rocks and integrates experimental findings with coupled modeling workflows to better understand injectivity, mechanical integrity, and long-term performance in lower-carbon subsurface energy systems.

My research focuses on developing MXene-based electrocatalysts for electrochemical energy conversion technologies, including hydrogen production and fuel cells. I investigate how composition, surface chemistry, and atomic structure influence catalytic activity and stability in hydrogen evolution, oxygen evolution, and oxygen reduction reactions. Current work explores titanium-based carbide, carbonitride, and nitride MXenes synthesized through minimal- or no-hydrofluoric acid routes to improve the safety and scalability of MXene production. By establishing composition–structure–property relationships, this research aims to provide design principles for advanced materials that support scalable, lower carbon-intensity energy technologies.

Eugenie Marie Pranada is a Ph.D. candidate in Materials Science and Engineering at Texas A&M University under the supervision of Dr. Abdoulaye Djire. She earned her M.S. in Materials Science and Engineering from Texas A&M University and her B.S. in Chemical Engineering from Mapúa University in the Philippines. Her research focuses on MXene-based electrocatalysts for sustainable energy applications, particularly hydrogen production and fuel cells, with an emphasis on understanding how surface chemistry and atomic structure govern catalytic performance. Her work has resulted in publications in journals including ACS Catalysis, Chem Catalysis, and Sustainable Energy & Fuels, and has been presented at numerous national and international conferences.

Power electronic converters are fundamental building blocks of modern energy systems, including renewable energy installations, electric vehicles, battery energy storage systems, and DC microgrids. My research focuses on the development of Conversion Function Theory (CFT), a novel framework for modeling and analyzing switching power converters and converter-dominated power systems.

CFT enables accurate and computationally efficient representation of converter behavior while preserving key dynamic characteristics of physical systems. The resulting models can be applied to stability assessment, control design, system optimization, and the integration of renewable energy and energy storage technologies. The long-term goal of this work is to provide scalable modeling tools that accelerate the design and deployment of next-generation electrified energy systems.

Ali Shawartamimi is a Ph.D. student in Electrical and Computer Engineering at Texas A&M University under the supervision of Professor Mehrdad Ehsani. He received his M.S. in Electrical Engineering from Texas A&M University in 2023 and his B.S. in Electrical Engineering from Palestine Polytechnic University in 2019, graduating at the top of his class.

His research focuses on power electronics, renewable energy systems, DC power distribution networks, and the development of Conversion Function Theory (CFT) for switching power converters. His work aims to develop accurate and computationally efficient modeling techniques for converter-dominated energy systems.

Hydraulic fracturing is the backbone of unconventional shale development, yet designing effective treatments remains one of the hardest problems in completion engineering. The numerical simulations used to model fracture growth and proppant transport are computationally expensive, often taking hours for a single run, so engineers cannot evaluate enough design alternatives to find genuinely optimal completions. My research addresses this gap by developing deep transfer learning surrogate models that generate two-dimensional maps of hydraulic fracture conductivity and proppant concentration at a fraction of the computational cost. These maps serve as direct inputs for optimizing completion design, allowing engineers to refine pumping schedules, fluid volumes, and proppant placement to maximize production and resource efficiency. The work scales from single-stage prediction toward multi-stage and pad-scale multi-well modeling, where field development decisions are actually made, and applies transfer learning so that a model trained in one basin can be adapted to other shale plays without retraining from scratch.

Oluwasanmi Talabi is a Ph.D. student in Petroleum Engineering at Texas A&M University, working under the supervision of Prof. Siddharth Misra. He holds a bachelor’s degree in Petroleum and Gas Engineering from the University of Lagos, Nigeria, and a master’s degree in Petroleum Engineering from Texas A&M University. His research develops deep learning surrogate models that replace expensive physics-based simulations with fast, reliable proxies for complex subsurface processes, with his doctoral work focused on hydraulic fracture prediction for unconventional shale development. He has broad experience applying machine learning across subsurface problems, including geological carbon storage and geothermal exploration. Oluwasanmi has held a research internship with Halliburton, and he is active in the Texas A&M SPE community, where he serves as Website Director for the student chapter.

Hydrogen storage remains a major bottleneck in hydrogen energy systems due to hydrogen’s inherently low volumetric energy density. Safe and efficient storage is essential not only to mitigate operational hazards, such as fires and explosions, but also to improve system profitability by maximizing performance, reducing energy consumption, and minimizing operational downtime. Metal hydrides present a promising alternative by enabling hydrogen storage at lower pressures and in solid form through adsorption. However, this adsorption process is highly exothermic and generates a substantial amount of heat, increasing the risk of thermal runaway. To address this, a safety-intelligent model predictive control (MPC) strategy was developed to maintain thermal stability in metal hydrides while maximizing storage efficiency. The controller was implemented on a microcontroller and validated experimentally under closed-loop conditions. Results demonstrate that the metal hydride system exhibits minimal temperature overshoot while achieving maximum hydrogen storage. These findings indicate that the safety-intelligent MPC effectively optimizes operation and has the potential to advance metal hydride storage. Overall, this work aligns with Chevron’s aim of advancing a lower-carbon H2 value chain through process intensification.

Zhane Ann Tizon is a third-year Ph.D. student in Chemical Engineering at Texas A&M University. She is working in the process systems engineering field under the supervision of Prof. Efstratios N. Pistikopoulos. Her research focuses on the modeling, safety, control, and operation of metal hydride hydrogen storage systems as part of the NSF RETRO Project. Her work on cyber-physical hydrogen storage led to her being named a co-inventor on U.S. Patent Application No. 63/951,033, “Real-Time Safety-Aware Control and Monitoring System Design.”

Her interest in sustainable energy systems originated during her undergraduate research at Mapúa University in the Philippines, where she earned her B.Sc. in Chemical Engineering. Building on this foundation, she continues to pursue research that advances safety and performance of emerging hydrogen storage technologies. She currently serves as Vice President of the Texas A&M Energy Research Society, where she advocates for and supports students pursuing research in the energy sector.

Dimitrios’s research focuses on advancing scalable decision-support workflows for enhanced geothermal systems and subsurface energy storage through hybrid reduced-order and physics-guided reservoir models. He develops integrated workflows to characterize coupled pressure, temperature, saturation, plume, and fracture-sensitive behavior under changing controls, permeability fields, and well configurations. His work supports critical components of safe and efficient deployment in EGS, CO₂ sequestration, and underground hydrogen storage, including injection and production optimization, pressure management, adaptive thermal management, monitoring, containment-risk assessment, and fracture/fault reactivation analysis. He also explores the use of data-driven reduced-order models, DMD variants, and scientific machine learning techniques to accelerate scenario evaluation and optimize subsurface operations while preserving the dominant physics needed for field-scale forecasting and decision-making.

Dimitrios Voulanas is a Ph.D. candidate and Graduate Research Assistant in Petroleum Engineering at Texas A&M University, working with Dr. Eduardo Gildin and Dr. George Moridis. He is affiliated with the Harold Vance Department of Petroleum Engineering and the Texas A&M Energy Institute. His doctoral research focuses on computational methods for subsurface energy systems, with emphasis on process-aware reduced-order modeling, dynamic mode decomposition, neural-network surrogate models, and optimization for reservoir-scale forecasting and decision-making.

His work is grounded in applied mathematics, numerical modeling, and optimization, with a focus on developing surrogate models that are not only fast, but also physically meaningful, stable, and useful for engineering decisions. A central part of his research is developing the theoretical and algorithmic foundations behind these models, including how reduced-order representations are constructed, how they preserve the dominant dynamics of reservoir systems, and how they can be integrated rigorously into optimization workflows. He is particularly interested in the mathematical structure of reduced-order models, the accuracy and robustness of data-driven approximations, and the validation of surrogate-based workflows against high-fidelity reservoir simulations.

His current research applications include CO₂ storage, enhanced oil recovery, geothermal energy, compressed gas storage, reservoir management, and broader energy-transition systems. Through this work, he aims to make complex simulation and optimization problems more practical for large-scale engineering applications while preserving decision-relevant accuracy and physical consistency. He is especially interested in methods that connect physics-based simulation, data-driven modeling, applied mathematics, and mathematical optimization to support more efficient, reliable, and environmentally responsible subsurface operations.

Dimitrios has an interdisciplinary background spanning geology, hydrogeology, petroleum engineering, reservoir simulation, applied mathematics, and scientific computing. He holds an M.S. in Petroleum Engineering from New Mexico Tech, an M.S. in Environmental Hydrogeology from Aristotle University of Thessaloniki, and a B.S. in Geology from Aristotle University of Thessaloniki. Before and during his graduate studies, he gained experience in wellsite geology, hydrogeological modeling, geothermal systems, GIS, reservoir characterization, multiphase flow simulation, and numerical modeling.

His broader research interests include robust optimization, scientific machine learning, uncertainty-aware reservoir management, numerical methods for subsurface flow, CO₂ utilization and storage, geothermal systems, and computational tools that translate advanced mathematical modeling into practical energy decisions.

As modern power grids increasingly rely on power electronics-interfaced generation, loads, and battery energy storage, maintaining system stability and operational reliability has become significantly more challenging, particularly under weak-grid and autonomous off-grid operating conditions. These challenges are further intensified by the highly dynamic characteristics of advanced AI computing loads, which can introduce strong interactions across multiple power electronic interfaces. My research develops AI-enabled coordination and control frameworks for power electronics-interfaced energy resources in modern power systems. A primary focus is on modeling and controlling the power electronic interfaces of emerging large loads, including AI data centers, crypto-mining facilities, and hydrogen electrolyzers. By integrating physics-based electromagnetic transient modeling with AI-driven adaptive control, my work enhances the grid-friendly integration of large loads. Ultimately, my research aims to enable scalable, resilient, and reliable coordination of large-scale power electronics-interfaced resources for next-generation AI computing infrastructure and electrified low-carbon industrial systems.

Xiaoyang Wang is a Ph.D. student in Electrical Engineering at Texas A&M University, advised by Prof. Xin Chen. Xiaoyang joined Texas A&M in Fall 2024. Prior to joining TAMU, he received both his bachelor’s and master’s degrees in Electrical Engineering from Xi’an Jiaotong University. His research focuses on dynamic modeling of large electronic loads, AI-enabled control, and the coordination of large-scale power electronics-interfaced resources in power systems. The open-source crypto-miner and AI data center models he developed in collaboration with ERCOT have been publicly released by ERCOT to support studies of large power-electronic loads. He has published multiple research papers in IEEE journals and conferences and has presented his work at academic seminars and workshops. He is also a recipient of the Hagler Institute for Advanced Study Graduate Fellowship and the Thomas Powell ’62 Fellowship.