A Generalized Derivative-Free Rate Allocation Optimization for Water and Gas Flooding Using Streamline-Based Method
- View All
Copyright 2017, Society of Petroleum Engineers. Rate allocation optimization for water and gas injection/production processes is typically complex, requiring multiple simulations to find the optimal reservoir management strategy for improved economic value of the asset. The objective of this paper is to develop and demonstrate a fast and robust derivative-free workflow which improves economic values via optimizing water and gas flooding rate allocation by streamline-based technique. Streamline-based rate allocation optimization has been demonstrated to be a powerful tool for application to waterflood operations. However, the utility of the technique has been limited in optimizing the Net-Present-Value (NPV) for improving the economics of these operations. In addition, the theoretical assumptions on physics limit its application to pressure-sensitive secondary or tertiary recovery processes such as gas injection. In the proposed workflow the expected NPV of each injector-producer pairs are evaluated at a given future business decision such as next infill time by using static, dynamic, and economic parameters such as price and discount rate along with time-of-flight. Then new flow rates are allocated based on the performance of the wells ranked by expected NPV to achieve better future economic value. The proposed workflow was first compared with previous streamline-based rate reallocation technique using series synthetic models. Although all tested methods showed better performance when compared with the base scenario, our proposed approach showed improved performance in terms of NPV, primarily due to proper handling of the reservoir dynamics and economic value in the objective function. The workflow was then benchmarked by a case study of a field subjected to waterflood and gas injection. The results of proposed approach were comparable to the population based derivative-free techniques such as Genetic Algorithm (GA) or Particle Swarm Optimization (PSO) where many simulations are required to achieve a similar outcome. The proposed workflow was coupled with next generation simulator and applied to various field studies. The method provides the ability to simultaneously control injector and producer flow rates for improving the economic value under multiple constraints. The workflow retains advantages of the conventional streamlinebased technique such as fast post processing and ease of application to a broad field studies.
author list (cited authors)
Tanaka, S., Kam, D., Xie, J., Wang, Z., Wen, X., Dehghani, K., Chen, H., & Datta-Gupta, A.