When transformers obey physics: A physics-guided transformer with symbolic regression for real time monitoring of flowing bottom-hole pressure
Engineering Applications of Artificial Intelligence · 2026
Abstract
Research topics
Related research
The Rock Physics Handbook
The Rock Physics Handbook addresses the relationships between geophysical observations and the underlying physical properties of rocks. It distills a vast quantity of background theory and laboratory results into a series of concise chapters that provide practical solutions to problems in geophysical data interpretation. This expanded second edition presents major new chapters on statistical rock physics and velocity-porosity-clay models for clastic sediments. Other new and expanded topics include anisotropic seismic signatures, borehole waves, models for fractured media, poroelastic models, and attenuation models. This new edition also provides an enhanced set of appendices with key empirical results, data tables, and an atlas of reservoir rock properties – extended to include carbonates, clays, gas hydrates, and heavy oils. Supported by a website hosting MATLAB routines for implementing the various rock physics formulas, this book is a vital resource for advanced students and university faculty, as well as petroleum industry geophysicists and engineers.
The Rock Physics Handbook
Responding to the latest developments in rock physics research, this popular reference book has been thoroughly updated while retaining its comprehensive coverage of the fundamental theory, concepts, and laboratory results. It brings together the vast literature from the field to address the relationships between geophysical observations and the underlying physical properties of Earth materials - including water, hydrocarbons, gases, minerals, rocks, ice, magma and methane hydrates. This third edition includes expanded coverage of topics such as effective medium models, viscoelasticity, attenuation, anisotropy, electrical-elastic cross relations, and highlights applications in unconventional reservoirs. Appendices have been enhanced with new materials and properties, while worked examples (supplemented by online datasets and MATLAB® codes) enable readers to implement the workflows and models in practice. This significantly revised edition will continue to be the go-to reference for students and researchers interested in rock physics, near-surface geophysics, seismology, and professionals in the oil and gas industries.
Characterization of complex fluvio–deltaic deposits in Northeast China using multi-modal machine learning fusion
Due to the lack of petroleum resources, stratigraphic reservoirs have become an important source of future discoveries. We describe a methodology for predicting reservoir sands from complex reservoir seismic data. Data analysis involves a bio-integrated framework called multi-modal machine learning fusion (MMMLF) based on neural networks. First, acoustic-related seismic attributes from post-stack seismic data were used to characterize the reservoirs. They enhanced the understanding of the structure and spatial distribution of petrophysical properties of lithostratigraphic reservoirs. The attributes were then classified as varied modal inputs into a central fusion engine for prediction. We applied the method to a dataset from Northeast China. Using seismic attributes and rock physics relationships as input data, MMMLF was performed to predict the spatial distribution of lithology in the Upper Guantao substrata. Despite the large scattering in the acoustic-related data properties, the proposed MMMLF methodology predicted the distribution of lithological properties through the gamma ray logs. Moreover, complex stratigraphic traps such as braided fluvial sandstones in the fluvio-deltaic deposits were delineated. These findings can have significant implications for future exploration and production in Northeast China and similar petroleum provinces around the world.