Efficiency prediction and scheme optimization on surfactant flooding based on neural network method and its application in the Ordos Basin

Abstract

The Ordos Basin is characterized by low-permeability oil and gas reservoirs. In the past, the recovery factors for such reservoirs were often predicted through laboratory experiments or numerical simulations, and methods for improving recovery were proposed based on reservoir characteristics. However, due to the differences among blocks, neither laboratory experiments nor numerical simulations have yet established a unified prediction standard for effect of oil flooding, and there is a lack of comparability between the production capacities of different blocks. In order to investigate the impact of different injection schemes on the oil flooding effect in the same type of reservoirs, this paper takes the Ordos Basin oil and gas reservoirs as an example and, considering the main factors controlling enhanced oil recovery, conducts calculations and analyses. A neural network-based method for predicting the recovery factors of different blocks is proposed, and the effects of injection volume, timing, concentration, and rate on the recovery factors are analyzed. The new method proposed in this study has important academic value for the subsequent optimization of enhanced oil recovery schemes in the same type of blocks.