Open AccessINTERWELL CONNECTIVITY AND DIAGNOSIS USING CORRELATION OF PRODUCTION AND INJECTION RATE DATA IN HYDROCARBON PRODUCTION
Author(s) -
Jerry L Jensen,
Larry W. Lake,
Thang D. Bui,
Ali Al-Yousef,
Pablo Hugo Gentil
Publication year - 2004
Language(s) - English
Resource type - Reports
DOI - 10.2172/833472
Subject(s) - estimator , ridge , regression , production (economics) , variance (accounting) , econometrics , statistics , field (mathematics) , oil field , extension (predicate logic) , regression analysis , computer science , geology , petroleum engineering , mathematics , paleontology , economics , accounting , pure mathematics , macroeconomics , programming language
This report details much of the progress on inferring interwell communication from well rate fluctuations. The goal of the project was to investigate the feasibility of inferring reservoir properties through weights derived from correlations between injection and production rates. We have focused on and accomplished the following items: (1) We have identified two possible causes for the source of negative weights. These are colinearity between injectors, and nonstationarity of be production data. (2) Colinearity has been addressed through ridge regression. Though there is much to be done here, such regression represents a trade-off between a minimum variance estimator and a biased estimator. (3) We have applied the ridge regression and the original Albertoni procedure to field data from the Magnus field. (4) The entire procedure (with several options) has been codified as a spreadsheet add-in. (5) Finally, we have begun, and report on, an extension of the method to predicting oil rates. Successful completion of these items will constitute the bulk of the final year's report