What is it about?
Integration techniques of well-log and seismic data (with their seismic attributes) for reservoir characterization are currently applied by the oil and gas industry for mapping rock physical properties. We have shown that the Artificial Neural Network method combined with well logs and seismic attributes predict density and volume of clay images at seismic scales. An algorithm was developed to convert the volume of clay image to a velocity image. We have applied this technique to a stacked P-wave reflected seismic section in the Tenerife field located in the Middle Magdalena Valley Basin in Colombia.
Featured Image
Photo by Donald Giannatti on Unsplash
Why is it important?
The results have been tested by creating synthetic seismograms with the rock physical properties predicted at seismic scale. The synthetics agree with the observed seismic data, which implies that the model is good. The velocity model that was produced was used as a first guess of an inversion algorithm.
Perspectives
The waveform inversion can provide velocity images with high degree of resolution that can delineate important geological features in hydrocarbon reservoirs. We hope the article makes the geophysical community aware of the importance of having realistic initial velocity model for inversion algorithms.
Geophysicist Jorge O Parra
Southwest Research Institute
Read the Original
This page is a summary of: Validated Artificial Neural Networks in determining petrophysical properties: A case study from Colombia, Interpretation, August 2018, Society of Exploration Geophysicists,
DOI: 10.1190/int-2018-0011.1.
You can read the full text:
Resources
Contributors
The following have contributed to this page
