Inversion Bandwidth Extension in Wara-Burgan Formation Using AI-ML Approach

Abstract Seismic inversion is crucial for extracting key parameters from seismic data, identifying geological composition, and characterizing reservoirs. Traditional methods, such as P-impedance and AVO inversion, face challenges due to data bandwidth limitations. This study integrates synthetic data-driven techniques with real data, including convolutional neural networks (CNN) and transfer learning, to improve seismic reservoir characterization. The study found that the CNN achieved nearly 94% prediction accuracy with low error rates, aligning with well data and demonstrating superior lateral continuity, even in blind well scenarios.