Petrel Automated Structural Interpretation—Ant Tracking

Depending on your workflow, Automatic Fault extraction can be performed at any scale. During the exploration phase, the focus may be to look for major tectonic fault systems spanning the basin and identify their influence on any prospects. Alternatively with known reservoirs, at appraisal, development or production phase, a similar approach can be applied, but at a local scale. A this scale you are looking at faults and fault systems that may be the result of tectonic forces from completely different directions which have affect on ultimate hydrocarbon recovery.

The Ant Tracking workflow consists of four independent steps:

1. Enhance the spatial discontinuities in your seismic data using any edge detection algorithm (i.e. variance, chaos, edge detection) and optionally, pre-condition your seismic data by reducing noise.
2. Generate the Ant Track Cube and extract the fault patches
3. Validate and edit the fault patches
4. Create final fault interpretation model

The first step in the workflow involves the reduction of noise in your seismic data and the creation of an edge-enhancing attribute (i.e. variance, chaos, dip deviation), highlighting special discontinuities.

During the second step, the Ant Track cube is created. The Ant Tracking algorithm follows an analogy of ants finding the shortest path between their nest and their food source by communicating using pheromones, a chemical substance that attracts other ants. The shortest path will be marked with more pheromones than the longest path and so the next ant is more likely to choose the shortest route, and so on.

The idea is to distribute a large number of these electronic "ants" in a seismic volume; and let each ant move along what appears to be a fault surface while emitting "pheromone." Ants deployed along a fault should be able to trace the fault surface for some distance before being terminated. Surfaces meeting expectations will be strongly marked by "pheromone." Surfaces unlikely to be faults will be unmarked or weakly marked. The Ant Tracking algorithm creates a new fault attribute highlighting the corresponding fault-surface features having orientations within some pre-determined settings. The algorithm then automatically extracts the result as a set of fault-patches, a highly detailed mapping of discontinuities.

Manual interaction is provided in the third step. The extracted fault patches must be evaluated, edited and filtered in order to obtain the final interpretation. This is done using an innovative approach applying an interactive stereo-net and histogram filter tool.

In the final step, the fault patches are used for further seismic interpretation or as input to the fault modeling directly.