Petrophysics
7 papers3 directions
Petrophysics is a research venue focused on the study of physical and chemical properties of rocks and their contained fluids. Between 2019 and 2023, the venue has published work that increasingly incorporates machine learning and artificial intelligence to address complex subsurface challenges. This research typically centers on the development of interpretation models and the processing of geophysical datasets, such as nuclear magnetic resonance (NMR) logging data and well-site tomography.
The venue is characterized by the application of diverse computational methods to geological problems. Published research includes the use of 1D convolutional neural networks and deep Q-learning for automated well-log depth matching, as well as symbolic regression for creating petrophysical interpretation models. Other studies focus on physics-driven deep learning networks for solving nonlinear inverse problems and the application of Bayesian inversion to optimize core analysis and formation testing procedures.
Petrophysics serves as a specialized forum for both academic and industrial researchers, with frequent contributions from organizations such as Halliburton and authors including Jiefu Chen, Xuqing Wu, and Songhua Chen. Within the broader machine learning community, the venue represents a domain-specific application area where data-driven techniques are integrated with physical principles to improve the accuracy and efficiency of formation evaluation and reservoir characterization.
Papers by Year
| 2023 | 1 |
| 2022 | 2 |
| 2021 | 2 |
| 2020 | 1 |
| 2019 | 1 |
Top Research Directions
Research directions most represented by papers from this venue.
| Multi-Component Loss Functions for Physics-Informed Learning | 1 |
| Generalizing One-Dimensional Models to Higher Dimensions | 1 |
| The Real World Validation Gap of Synthetic Data | 1 |
Top Authors
| Author | Papers |
|---|---|
| Xuqing Wu | 2 |
Impactful Papers
Papers from this venue that shaped research clusters.
Foundational
A Physics-Driven Deep-Learning Network for Solving Nonlinear Inverse Problems
foundational
Multi-Component Loss Functions for Physi
2020