Transforming the subsurface energy landscape—encompassing both geothermal and oil and gas—relies on innovative solutions like artificial intelligence (AI) and electrification to drive true sustainability.
AI is more than just a buzzword; it is a transformative force reshaping energy production, making it cleaner, safer, and more efficient. By enabling accurate predictions, automating processes, reducing human intervention, and streamlining workflows, AI significantly enhances resource management and real-time decision-making.
Central to AI’s success are five foundational elements: standardization, automation, integration, scalability, and continuous improvement. Together, these elements provide a clear roadmap for effective AI deployment, addressing challenges and driving sustainability across the subsurface energy sector.
- Standardization is the bedrock of AI implementation, ensuring consistent, high-quality data that allows AI to accurately identify drilling hazards, optimize well placement, and assess geothermal reservoirs. This reliable data generates actionable insights, reducing risks and improving safety.
- Automation elevates efficiency further by managing routine tasks, enabling real-time adjustments, and minimizing nonproductive time. It reduces human exposure to hazardous situations, enabling engineers to focus on complex, high-value tasks.
- Integration is the critical link that ensures seamless data flow across teams and departments, fostering collaboration and enabling swift, informed decision-making. This interconnectedness paves the way for smarter, more cohesive operations.
- Scalability allows AI to adapt to increasing data volumes, delivering insights that span multiple wells and geothermal sites while ensuring consistent performance and efficiency.
- Continuous improvement keeps AI models relevant and accurate, evolving to predict equipment failures, refine production schedules, and optimize geothermal management as operational needs shift.
This transformation goes beyond mere efficiency; it reimagines how we explore, produce, and sustain energy for a smarter, safer, and more adaptable future. The subsurface energy sector is inherently data-intensive, with vast amounts of information gathered from seismic surveys, drilling operations, reservoir simulations, and production monitoring. However, inconsistent and variable data can present significant challenges. Standardization addresses this issue by enforcing uniform data structures across sources like well logs, core samples, and production reports. It also encourages adopting advanced analysis techniques that move beyond traditional simplifying assumptions.
Research teams at the University of Houston are at the forefront of this transformation, demonstrating that AI is not just a tool but a catalyst for change in the subsurface energy industry. Through projects like the use of wavelet transform, researchers are developing general solutions to analyze recorded signals effectively, marking the first step in AI implementation—standardization. This work paves the way for broader AI integration, ultimately leading to full-scale implementation across the subsurface energy sector.
In one example, wavelet transform projects have shown that physical events like fracture closure can be accurately identified using the continuous wavelet transform. The results were validated by independent measurements such as strain gauges. This represents a significant advancement in accurately detecting challenging physical events, a topic debated in the oil and gas industry since 1978.
This approach not only facilitates timely detection of physical events but also introduces an innovative, generic analysis technique for identifying dynamic fracture propagation using available data, such as treating pressure (SPE 217789). It has been validated with independent measurements, including the development of microseismic events over time. Standardization through the continuous wavelet transform enabled the automation of predicting the microseismic cloud from treating pressure, which effectively determines the stimulated reservoir volume in unconventional reservoirs. This allows AI to interpret fracture events within the hydraulic fracture system and translate them into a microseismic cloud—a process that is typically costly to record during each treatment in the oil and gas industry.
The new approach introduces effective machine learning (ML) modeling for hydraulic fracturing operations. The integration of different data sources provides a cost-effective solution. Furthermore, this technique can be extended to geothermal reservoir applications to estimate the activated reservoir volume (ARV), optimizing the size of enhanced geothermal systems (EGS). ML models can be fine-tuned further through continuous improvement.
The introduction of wavelet transforms, particularly the continuous wavelet transform, in subsurface energy applications has provided game-changing analyses of physical events using signal processing techniques that can be generalized over several applications. This advancement has proven impactful not only in theoretical modeling but also in practical field operations, such as the automation of fracture event detection during hydraulic fracturing. It facilitates more precise and effective operational management, bridging the gap between theory and real-world practice. By enhancing decision-making and operational efficiency, it offers a comprehensive, real-world example of AI implementation in the subsurface energy sector. Additionally, wavelet transform stands as a success story of complete AI implementation within this field.
Meanwhile, electrification complements AI by reducing reliance on traditional fuels, lowering emissions, and supporting net-zero goals. Electrification of well stimulation techniques represents a significant step toward a cleaner, more sustainable energy future. Traditionally, well stimulation in oil, gas, and geothermal reservoirs relies heavily on hydraulic fracturing and acid stimulation techniques that consume vast amounts of water and special chemicals that may not be environmentally safe. By shifting to electrical stimulation techniques like waterless fracturing using plasma pulse, the environmental impact is reduced—minimizing carbon emissions, noise pollution, and water usage.
Although still in the research phase, the project has shown great potential and exceptional results that could replace conventional chemical stimulation techniques removing its harmful environmental impacts like chemicals wastes. Beyond engineering innovation, it is a commitment to future generations, ensuring we meet energy needs while caring for the planet. This game-changing technology introduces a human element, promoting safer, more environmentally conscious energy practices that benefit both communities and ecosystems.
Wavelet transform, once confined primarily to mathematical texts, medical applications, and electrical engineering, has undergone a remarkable evolution in subsurface energy applications. Driven by UH’s Mohamed Soliman’s pioneering efforts, its use has expanded significantly in well-test analysis and fracturing over the past 21 years, while his innovative plasma stimulation research has seen over 15 years of continuous development. This extensive work, reflecting decades of research and refinement, has brought these technologies to the brink of practical field deployment. Grounded in years of expertise within the oil and gas sector, these advancements have established a robust foundation, making the techniques not only more efficient and reliable but also more impactful across both conventional and unconventional reservoirs. These research breakthroughs represent confident strides toward field implementation and global commercialization, paving the way for scalable, cost-effective solutions that align with the evolving demands of the energy sector.
University of Houston research teams continue to deliver innovative solutions that drive the energy industry toward cleaner and more sustainable practices. With decades of development, these advancements have paved the way for field-ready applications that optimize energy extraction while minimizing environmental impact.
Despite this progress, the success of these innovations depends heavily on adequate funding. Investment in research, technology development, and field applications is essential to bring these advancements to full-scale implementation. By supporting these projects, stakeholders can accelerate the transition toward cleaner energy production, drive long-term sustainability, and ensure the effective deployment of AI and electrification across subsurface energy sectors. Continued financial support will be pivotal in advancing these technologies, allowing them to evolve further and make a tangible impact on global energy goals.
For Further Reading
A New Technique for Estimating Stress from Fracture Injection Tests Using Continuous Wavelet Transform by M. Gabry, I. Eltaleb, M. Soliman, and S. Farouq-Ali, University of Houston.
Validation of Estimating Stress from Fracture Injection Tests Using Continuous Wavelet Transform with Experimental Data by M. Gabry, I. Eltaleb, M. Soliman, and S. Farouq-Ali, University of Houston.
Hydraulic Fracture Closure Detection Techniques: A Comprehensive Review by M. Gabry, I. Eltaleb, A. Ramadan, A. Rezaei, and M. Soliman, University of Houston.
Calibration of Continuous Wavelet Transform for Dynamic Hydraulic Fracture Propagation with Micro-Seismic Data: Field Investigation by M. Gabry, M. Soliman, S. Farouq-Ali, Eltaleb; University of Houston, C. Cipolla; Hess Corp., A. Gharieb; Apache Egypt Corp.
Advanced Deep Learning for Microseismic Events Prediction for Hydraulic Fracture Treatment via Continuous Wavelet Transform by M. Gabry, M. Soliman, S. Farouq-Ali, I. Eltaleb; University of Houston, C. Cipolla; Hess Corp., A. Gharieb; Apache Egypt Corp.
Waterless Fracturing Using Plasma Pulse, Mohamed Soliman
Pulse Power Plasma Stimulation: A Technique for Waterless Fracturing, Enhancing the Near-Wellbore Permeability, and Increasing the EUR of Unconventional Reservoirs by M. Gabry, A. Rezai, M. Khalaf; P. Gordon, ExxonMobil; University of Houston, C. Cipolla, Hess Corp.
