Design of Smart Wellhead Controllers for Optimal Fluid Injection Policy and Producibility in Petroleum Reservoirs: A Neural Network-Model Predictive Approach
Design of Smart Wellhead Controllers for Optimal Fluid Injection Policy and Producibility in Petroleum Reservoirs: A Neural Network-Model Predictive Approach
byMasoud Nikravesh, Masoud Soroush, M. R. Johnston, Tadeusz W. Patzek
Year:1997
Bibliography
Nikravesh M., M. Soroush, M. Johnston and T. W. Patzek, Paper SPE 37445, “Design of Smart Wellhead Controllers for Optimal Fluid Injection Policy and Producibility in Petroleum Reservoirs: A Neural Network-Model Predictive Approach,” presented at the SPE Production Operations Symposium, March 9-11, 1997, Oklahoma City. OK.
Abstract
In this paper we present the second generation of “smart” controllers for fluid injection and predictive techniques for developing optimal fluid injection policy. A neural network model is used to optimize the oil field management in terms of fluid injection and oil recovery. We illustrate performance of the neural network for model identification and control purposes. In particular, the neural network models are used to control and predict the behavior of individual injectors in CalResources’ Phase III steam drive pilot in the South Belridge Diatomite, CA, under different injection policies.
