Using linear–quadratic regulator to optimally control the gas lift operation

Abstract

As the reservoir pressure declines, production decreases and there is no way just using artificial lift methods such as the gas lift to make the production economical. Gas lift optimal control is one of the dominant problems in gas lift operation. There are various methods for this purpose in literature, but all of them use a primary control method with a simple and inaccurate simulation model which does not guarantee enough accuracy. Besides, in all of the previous studies, the control model was just concentrated on well, while the part that converts a static optimization to an optimal control one is the reservoir and the main constraints of the optimization problem are on the surface facilities. Here, first using simulation software an integrated model of reservoir, wells, and surface facilities are developed, then for that, an accurate proxy model is created and using the linear–quadratic regulator (LQR), the optimal trajectory of the lift gas injections for all wells are calculated. Finally, the results are validated by comparison with two other heuristic optimization algorithms, and results confirmed the better performance of the method of this paper in comparison with two heuristic methods, higher net present value (NPV) and a highly stable control trajectory of the method of this paper.