The Capacitance Resistance Model (CRM) is a data driven method for characterizing a reservoir and optimizing its oil production without complex and time-consuming reservoir simulations or need for geological modeling.
This method has been developed over eight years of research at The University of Texas at Austin under the supervision of Dr. Larry W. Lake and has been successfully applied in the industry.
CRM requires only historical rate data that is collected in most waterfloods. The more data the better. The data used are water production rates by well, oil production rates by well, water injection rates by well and well coordinates for both injection and production wells. If available, bottom hole pressures for the producing wells can also be used.
The method has been implemented in software that consists of three modules that can perform the characterization and optimization in one continuous run. The modules are the Total Production Model Fit, Oil Production Model Fit and Optimizer. Each run typically lasts 1-2 hours for most reservoirs.
The Total Production Model Fit module manipulates injector-producer connectivities and producer time constants to obtain the best possible match between the total production predicted and the total production measured over time for each well. The outputs of this module are the injector-producer connectivities and the producer time constants. This, the inferred geology of the reservoir, can be used to optimize the injection distribution into the future.
The Oil Production Model Fit module fits each producer’s oil production over time using two empirical constants that are used in the Optimizer along with the connectivities and time constants to predict future production.
Once the best fits are obtained, the Optimizer will use economic parameters, maximum allowed injection rate for each injector, maximum allowed total injection rate, and an optimization time horizon to calculate the optimum injection rate for each injector for the future. Optimum is defined as the set of conditions that produce the maximum net present value (NPV) over a pre-specified future interval.
The following bar chart shows a typical result from the optimization: injector shares before and after the optimization. You can tell at a glance how the injection water should be redistributed to increase the performance of the reservoir.
The following plot shows the estimated impact of the reallocation of injection on the oil production. It compares the cumulative oil production using the optimum injection scheme versus the cumulative oil production in the future using the current (base) injection scheme.