For capital-intensive exploration and production projects, understanding the chemical and physical properties, phase behavior, compatibility, spatial distribution, and hydraulic connectivity of reservoir fluids is critical for long-term planning and operation. Often, reservoir fluid samples acquired via formation testing represent the only source of fluid properties reliable enough for economic evaluation. Therefore, the ability to consistently collect representative high-quality samples is essential for successful exploration and appraisal projects.
Current methods for obtaining reservoir fluid are complicated by contamination with mud-filtrate, which causes error in lab measurements and makes the data unreliable. Optimally timing when samples are taken is key to avoid contaminants and provide clean formation fluid samples.
Researchers at the University of Texas at Austin have developed a novel method for performing automated, real-time optimization of formation-tester measurements to improve sample quality and consistency while simultaneously reducing the time required to obtain such samples. The proposed technology uses artificial intelligence to monitor downhole measurements and automatically optimize formation-tester operating parameters during cleanup and sampling operations. Artificial intelligence is leveraged to estimate real-time mud-filtrate contamination during pump-out, predict how contamination will vary as pumping progresses, alter operating parameters to optimize mud-filtrate cleanup, and identify potential downhole issues, such as probe plugging. This stand alone algorithm can be used with existing wireline and while-drilling tools.
We are currently seeking commercial partners to test this technology in the field. A US provisional patent application has been filed on this technology.