Seismic Fault Enhancement Based on Multi-directional Component Analysis

Session: Faults, Fractures, and Geomechanics for the Energy Transition

Presenting Author:

Authors:

Abstract:

The discontinuity attribute volume of post-stacked seismic data can reveal the spatial characteristics of faults to a certain extent. Directly using discontinuity attribute volumes to identify faults is often susceptible to noise interference, resulting in poor detection performance for complex fractures. Conventional fault enhancement processing algorithms process discontinuity attribute volumes based on the constraints of unidirectional information of local faults, suppressing noise interference in non-fault directions. Their computational accuracy depends on the accuracy of fault direction calculation, and there are significant errors at fault intersection locations. In the fault enhancement method based on multi-directional component analysis, the local fault signals of discontinuity attribute data are regarded as a mixture of plane wave signals in multiple different directions. Multi-directional component analysis is used to decompose the mixed signals in different directions into multiple unidirectional plane wave signals, and select and retain the effective plane wave signals, which can remove noise interference while preserving the cross-characteristics of fault components in different directions and enhancing the fault signals. When applied to actual seismic data, it can effectively improve the signal strength of weak signal faults, enhance the continuity of small fault signals, optimize the smoothness of main large faults, and correct the local cross-relationship of faults in different directions.