Unsupervised AI integration toward an efficient timeline reduction in a successful CCUS planning process.

Seismic data remains a pillar of subsurface modeling and the understanding of the potential for transitioning from oil and gas exploration to CCUS development planning and execution.

An artificial Intelligence based on genetic algorithm (GA) has been used successfully to automatically compute an extensive horizons/faults/attributes database.

After a processing step which lasted less than 12 minutes for each seismic volume, maps with various attributes were displayed for all the layers computed by the AI including amplitude for instance. Main faults planes and subtle features were also extracted and integrated into a database with their own properties and fault throw maps.

The genetic algorithm automatically generated a suites of e waveform, attributes and other characterization of surfaces and faults that help build stratigraphic /structural domain and seismic facies maps within the entire potential area top to bottom in less than a day.

Such an automatic, extremely fast, and unbiased approach can help the geoscientists directly focus their time and attention visualizing and interpreting the significance of the results delivered by the artificial intelligence for various applications.

This AI workflow will efficiently reduce the timeline and increase decision accuracy during the CCUS project selection and evaluation.