Subtle Fault Prediction Technique Based on the Integration of Deep Learning and Seismic Spectral Decomposition

Abstract

Faults often control the movement and aggregation of oil and gas. With the development of oil fields, the role of subtle faults is becoming more and more important. The accuracy of fault interpretation directly affects the direction of exploration and development. However, due to the limitation of the seismic resolution, it is hard to identify these faults according to routine methods such as coherence, variance, curvature, etc. To overcome such kind of challenge and better match the demand for fine fault identification, a method integrated deep learning and spectral decomposition was proposed.