Abstract
The problem of effective oil reservoir managing becomes quite challenging during the late stages of its production. That's why, going for the intermittent operational mode is typical use-case for low-rate wells with electrical submersible pump since it's more efficient from both economical and engineering point of view. So, finding optimal parameters becomes crucial for the getting best value from the changes made. Tackling this problem leads to creation of the complex algorithm which will allow to compose an accurate physically motivated mathematical representation of a real-world object and provide functionality of calculating the best-possible operating regime.
The physically driven mathematical model was designed which represents system of three key elements: "tubing–annulus–drainage area". They are connected with each other through boundary conditions defined in the submersible pump intake by mass conservation equations.
The optimization task can be split into two blocks. The first stage, adjusting the model to the real operational data to match calculated dynamics, is handled by introducing "adaptation" parameters which represents specifics of a certain tubing system and varying IPR curve.
The second stage, optimization itself, is implemented by various optimization-aimed algorithms which objective is maximizing income while following operational and geological constraints.
As the result the physically motivated model of well operated in intermittent mode was obtained. The program module was developed with use of designed model. This software provides functionality to predict operational dynamics of production, hydraulic and electrical behavior. The accuracy of this approach to representing periodic well was verified by close analysis of convergence with well-trusted transient flow oil and gas simulator.
Developed modeling tools were the core for optimization program which are developed for the sake of finding the best possible parameters for a well with an ESP pump operated in periodic mode. It is worth to mention that this algorithm is also takes into account all geological and mechanical constraints.
