Streamline-Based Transport Tomography With Distributed Water Arrival Times
Academic Article
Overview
Research
Identity
Additional Document Info
Other
View All
Overview
abstract
SummaryTraditional history matching involves calibration of reservoir models by use of well response such as production or tracer data aggregated during multiple producing intervals. With the advent of novel tracer technologies, we now can obtain distributed water or tracer arrival-time information along the length of horizontal or vertical wellbores. This provides significantly improved flow resolution for detailed reservoir characterization through inversion of distributed water or tracer arrival times in a manner analogous to travel tomography in geophysics.In this paper, we present an efficient approach to incorporate novel tracer-surveillance data and distributed water arrival-time information during history matching of high-resolution reservoir models. Our approach relies on a streamline-based work flow that analytically computes the sensitivity of water-arrival times with respect to reservoir heterogeneity, specifically porosity and permeability variations. The sensitivities relate the changes in arrival time to small perturbations in reservoir properties and can be obtained efficiently with the streamline-based approach with a single flow simulation. This makes the approach particularly well-suited for high-resolution reservoir characterization. Finally, the sensitivities are used in conjunction with an iterative inversion algorithm to update the reservoir models with existing and proven techniques from seismic tomography.The power and utility of our proposed approach are demonstrated with both synthetic and field examples. These include the SPE benchmark Brugge field case and an offshore field in North America. Compared with traditional history-matching techniques, the proposed tomographic approach is shown to result in improved resolution of heterogeneity through matching of water-arrival time at individual completions in addition to the aggregated well-production response. This results in improved performance predictions and better identification of bypassed oil for infill targeting and enhanced-oil-recovery (EOR) applications.
