Full Record

New Search | Similar Records

Title Extraction of useful signals from noise using advanced time-frequency analysis in Enhanced Oil Recovery
Publication Date
Date Accessioned
Discipline/Department Engineering Mechanics
University/Publisher University of Texas – Austin
Abstract In Enhanced Oil Recovery (EOR), the injected CO2 does not distribute evenly through ground layers and often is not going to the desired direction. The key issue of this research, was to extract the useful signal from noise for a novel CO2 subsurface imaging solution which uses surfaced based sensors only, in comparison with conventional seismic technology which requires downhole equipment [5, 7, 8, 19, 20]. That could potentially reduce the cost drastically due to its simplicity in installation, performance, less labor intensive and faster process of data. The extremely low signal-to-noise (S/N) ratio required the utilization of binomial time-frequency domains (TFDs) to process the collected field data as the problem involves extremely non-stationary signals. Binomial Cone-Kernel function is arguably the most advanced signal independent kernel; it allowed us to extract useful signals embedded in noise and observe repeatable waves for the first time.
Subjects/Keywords Tube waves; Guided waves; Time-frequency domains; Enhanced oil recovery
Contributors Lu, Nanshu (advisor); Djurdjanovic, Dragan (advisor)
Language en
Country of Publication us
Record ID handle:2152/63954
Repository texas
Date Indexed 2018-10-22
Issued Date 2017-08-04 00:00:00
Note [department] Engineering Mechanics;

Sample Images | Cited Works