Researchers have developed a novel hybrid analytical-physics-informed neural network (PINN) model for simulating subsurface heat transfer in geothermal systems. This approach combines analytical solutions with PINNs to efficiently handle the complexities of heterogeneous soil conditions and borehole heat exchangers. The model effectively removes singularity issues and captures bulk heat transfer by focusing on a learned correction to an idealized homogeneous approximation, demonstrating effectiveness across various analytical models. AI
IMPACT This hybrid model could improve the efficiency and accuracy of simulations for geothermal energy systems, potentially accelerating their development and deployment.
RANK_REASON The cluster contains a research paper published on arXiv detailing a new modeling technique.
- alphaXiv
- arXiv
- CatalyzeX
- cs.LG
- DagsHub
- Gotit.pub
- Hugging Face
- physics-informed neural networks
- Smajil Halilovic
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
