Interpretable Machine Learning for High-Accuracy Reservoir Temperature Prediction in Geothermal Energy Systems
Problem
Accurate prediction of reservoir temperature in geothermal energy systems is critical for optimization, but complex geothermal data challenges traditional modeling.
Why it matters
Reservoir temperature estimates directly affect how geothermal energy systems are optimized.
What the AI did
Compared SVR, random forest, Gaussian process regression, DNN and GNN models for reservoir temperature estimation, using ALE and SHAP for interpretability and key predictor identification.
Results
Random forest achieved lowest MSE (66.16) and highest R2 (0.977).
Impact
Reports a high-accuracy interpretable predictor for geothermal reservoir temperature in a single study.
Record
| Venue | Energies |
|---|---|
| Published | 2025 |
| Work type | article |
| Authors | Mohammad Ali Ahmadi |
| Citations | 9 (OpenAlex, at import time) |
| References | 45 |
| Model or method | SVR, random forest, Gaussian process regression, DNN, GNN; random forest best |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3390/en18133366 |
| Open access | gold |
| Keywords | geothermal, reservoir temperature, random forest, machine learning, SHAP |
Importance rating
real but narrower advance
Geothermal reservoir temperature prediction aids system optimization, but abstract gives no scale or impact evidence beyond a single comparative modeling study.
More in Engineering & Materials
| # | Paper | Year | Citations | Importance |
|---|---|---|---|---|
| 1 | Scaling deep learning for materials discovery | 2023 | 1,471 |
8
|
| 2 | Application of Machine Learning in Material Synthesis and Property Prediction | 2023 | 165 |
7
|
| 3 | Machine learning-guided discovery of ionic polymer electrolytes for lithium metal batteries | 2023 | 115 |
7
|
| 4 | Machine Learning for Structural Health Monitoring of Aerospace Structures: A Review | 2025 | 44 |
7
|
| 5 | Power System Transient Stability Assessment Based on Machine Learning Algorithms and Grid Topology | 2023 | 42 |
7
|