![[2506.17344] FFINO: Factorized Fourier Improved Neural Operator for Modeling Multiphase Flow in Underground Hydrogen Storage](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2506.17344] FFINO: Factorized Fourier Improved Neural Operator for Modeling Multiphase Flow in Underground Hydrogen Storage
Summary
The paper presents FFINO, a novel neural operator for modeling multiphase flow in underground hydrogen storage, demonstrating significant improvements in efficiency and accuracy over existing models.
Why It Matters
As the world transitions to a low-carbon economy, efficient hydrogen storage solutions are critical. This research offers a faster and more accurate modeling approach, which can enhance the management of underground hydrogen storage, making it a vital contribution to energy sustainability efforts.
Key Takeaways
- FFINO model reduces trainable parameters by 38.1% and training time by 17.6%.
- Achieves 9.8% accuracy improvement in hydrogen plume prediction.
- Inference time is 7,850 times faster than traditional numerical simulators.
- Injection rate is identified as the most influential parameter affecting model performance.
- The model serves as a stable alternative for real-time UHS applications.
Computer Science > Machine Learning arXiv:2506.17344 (cs) [Submitted on 19 Jun 2025 (v1), last revised 24 Feb 2026 (this version, v2)] Title:FFINO: Factorized Fourier Improved Neural Operator for Modeling Multiphase Flow in Underground Hydrogen Storage Authors:Tao Wang, Hewei Tang View a PDF of the paper titled FFINO: Factorized Fourier Improved Neural Operator for Modeling Multiphase Flow in Underground Hydrogen Storage, by Tao Wang and 1 other authors View PDF Abstract:Underground hydrogen storage (UHS) is a promising energy storage option for the current energy transition to a low-carbon economy. Fast modeling of hydrogen plume migration and pressure field evolution is crucial for UHS field management. In this study, a new neural operator architecture, factorized Fourier improved neural operator or FFINO is proposed as a fast surrogate model for multiphase flow problems in UHS. Experimental relative permeability curves reported in the literature are also parameterized as key uncertainty parameters for the FFINO model. FFINO model performance with the state-of-the-art Fourier-enhanced multiple-input neural operators or FMIONet model are systematically studied through a comprehensive combination of metrics. Our new FFINO model has 38.1% fewer trainable parameters, 17.6% less training time, and 12% less GPU memory cost compared to FMIONet. The FFINO model also achieves a 9.8% accuracy improvement in predicting hydrogen plume in focused areas, and 16.3% higher accuracy in p...
