Cost-Engineering Waterflooding Management Methods

Abstract

Optimization of the operation costs of Oil Companies in Western Siberia is the most important task of monitoring the development of oil fields. This is due to both: a decline in oil prices and an increase in the water cut in the production. Companies are forced to have large expenses associated with the organization of injection of the working agent to the reservoir pressure maintenance system, fluid lifting to the surface and operations on fluid dehydration. Often, the total value of operation costs forces companies to abandon the operation of wells, which negatively affects both the company's income and the level of oil production. The development of modeling tools opens up opportunities for companies to optimize key technologic and economic indicators of field development. This is especially relevant for old fields that are at the final stage of development, when the achievement of cost-effectiveness is impossible without constant optimization workovers. However, the geological uncertainties and the complexity of the correct evaluation of the reservoir simulation connection between the injection and production wells do not allow oil companies to receive a confident answer to the question of the efficiency of the current waterflooding system and individual injection wells. Unfortunately, the complexity of creating a permanent simulation model, which is connected both with the unreliability of input data, and with high labor and computational costs, does not allow to fully meet the requirements for optimizing the waterflooding system. At the same time, analytical methods, for instance, block-factor analysis (BFA) [1] despite its simplicity and flexibility, is not popular among reservoir engineers due to low prediction ability. In this regard, there is a need to create a new engineering tool, which would simultaneously have a good predictive ability and would be rapid in use. Tools that meet these requirements include, for instance, analytical models that use database mining (data-driven methods). The use of these models allow estimate the value of the hydrodynamic connection between the producing and injection wells, make a retrospective analysis of the waterflooding system and make a reliable forecast of the production change when the injection changes using and tuning on history of operation modes of the wells. The paper considers a hybrid reservoir simulation model based on the capacitance-resistive model (CM / capacitance-resistive model, CRM). The use of this model is based on training on history data, then testing the quality of training on test history data and subsequent forecasting development parameters. Based on physical processes, a simplified model of material balance with a minimum number of unknowns makes it possible to effectively predict the effect of injection wells parameters change. Also, this method allows, indirectly, qualitatively identify injectivity wells with unproductive withdrawal and, as a consequence, with a low production effect. The approach described in the paper, called the cost-BFA, is integration of the CR method and the economic model that allows to predict additional production, to minimize operation costs and to maximize net present value (NPV) taken into consideration the operational costs of the Company.