A New Qualitative and Quantitative Analytics Approach on Waterflood Operations Data for Improving Oil Recovery
A. Venkatraman, A. Malkov, A. Yadav, D. Davudov, K. Awemo, T. Hag and X. Chen
Event name: IOR 2019 – 20th European Symposium on Improved Oil Recovery
Session: Coffee Break / Poster Session Day 2
Publication date: 08 April 2019
Info: Extended abstract, PDF ( 1.13Mb )
Price: € 20
Water flooding is an established method of secondary recovery to increase oil production. While previous research has focused on designing waterflood operations, there are no tools to evaluate the efficacy of those designs and optimize it frequently based on data available during the course of water flooding operations. In this research, we present a novel approach of using data mining techniques to increase oil recovery using operations data from a field undergoing water flooding. The results presented in this research can be adapted to any field to optimize recovery at frequent intervals, where injection and production data is continuously available. Operations data from a current water flooding field is used to improve water injection strategy by using a combination of qualitative (cross correlation analysis) and quantitative analysis (capacitance resistance model). Field data obtained from each injector and the surrounding producers are used for cross correlation analysis that enable identifying thief zones. The qualitative insights obtained from the cross-correlation analysis are used to improve the capacitance resistance model for the field. The improved capacitance resistance model is used to obtain redistribution of water among injectors with the purpose of increasing oil recovery. Reservoir simulation prediction of oil recovery on the two cases (the previous benchmark case and the new optimized injection strategy obtained using data mining techniques) is presented. It can be seen that the redistribution of water obtained using this novel approach improves oil estimates in the range of 5-10%. A field case of using data mining techniques of cross correlation analysis and capacitance resistance modeling is presented as a means to improve reservoir characterization using operations data. The insights obtained by using a combination of these two methods are used to redistribute water injection in a producing field. This new approach can be used to optimize water injection at frequent intervals based on the operations data obtained from the field. Operational challenges in implementing redistribution of water injection rates frequently are highlighted for the sake of other operators implementing such an approach.