Accurate prediction of geothermal reservoir temperatures is crucial for the exploration and development of geothermal resources. In our previous work, we employed machine learning models, including deep neural networks (DNNs), to predict reservoir temperatures based on hydro-geochemical data from Western Anatolia's geothermal systems. While the DNN model outperformed traditional regression approaches, the limited size and variability of the dataset constrained the model's predictive capabilities. In this study, we integrate Generative Adversarial Networks (GANs) to generate synthetic hydro-geochemical data, effectively augmenting the existing dataset. By enhancing the training data with realistic synthetic samples, we aim to improve the performance and generalization of machine learning models for reservoir temperature prediction. Our results demonstrate that the GAN-augmented models achieve higher accuracy and lower error rates compared to models trained on the original dataset, offering a novel approach to address data scarcity in geothermal exploration.

Press the Back button in your browser, or search again.

Copyright 2025, Stanford Geothermal Program: Readers who download papers from this site should honor the copyright of the original authors and may not copy or distribute the work further without the permission of the original publisher.