Abstract

Waveforms acquired from acoustic well logs are essential for determining formation elastic properties used in subsurface energy applications such as carbon storage and geothermal energy, as well as for well monitoring and integrity assessment. Traditional processing relies on extracting acoustic modes and inverse modeling, which can be sensitive to noise and computationally intensive.

In this talk, I present two machine learning-based approaches to predict elastic properties directly from acoustic data. The first uses neural networks on dispersive modes (flexural and quadrupole slowness), enabling accurate and fast estimation of velocities in isotropic and VTI formations using limited data. The second approach uses convolutional neural networks (CNNs) to predict properties directly from raw waveforms, eliminating conventional processing while maintaining high accuracy and robustness to noise.

These approaches enable fast, reliable prediction of formation properties and support real-time subsurface characterization for energy transition applications.

Biography

Dr. Elsa Maalouf is an Assistant Professor in the Department of Chemical Engineering and Advanced Energy at the American University of Beirut, Lebanon. She earned her Ph.D. in Petroleum and Geosystems Engineering from University of Texas at Austin, USA. Her research focuses on sustainable subsurface energy applications, including geothermal energy, carbon storage, and the safe production of underground resources, with expertise in acoustics, geomechanics, and numerical modeling. Her work integrates physics-based understanding with data-driven approaches to advance reservoir characterization and monitoring.

More broadly, her research addresses sustainable materials and energy practices across the full life cycle, with an emphasis on improving energy efficiency through informed material selection, design, and performance optimization.

She was elected as the 2024–2026 SPWLA Regional Director for the Middle East and Africa and was selected in 2023 as a Future Leader of the American Rock Mechanics Association. She is also the recipient of the 2025 Society of Petroleum Engineers Regional Award in Formation Evaluation.