基于机器学习的角闪石温压计及其地质应用

Key Points

• The aim is to develop a temperature-pressure prediction model based on the composition of hornblende to better constrain magma crystallization conditions.
• Developed a prediction model using machine learning algorithms based on hornblende composition.
• Utilized a large dataset for training to assess the model's performance.
• Applied the model to various geological samples across different regions.
• The model effectively captures the complex nonlinear relationship between hornblende composition and temperature-pressure conditions.
• Predictions reveal depth and temperature distribution patterns of magma crystallization in different regions.
• Model exhibits broader applicability compared to traditional empirical formulas.