Well rates and location optimization considering genetic algorithms and surrogate models
Palavras-chave:
optimization, surrogate models, genetic algorithm, well placement and flow rate, reservoir engineeringResumo
In this work, we solve an optimization problem of a known reservoir of the literature through an
integrated optimization by a Genetic Algorithm (GA). The location of the wells and their flow rates are the
variables. The main objective of this paper is to maximize the net present value (NPV). The optimization utilized
the GA from Toolbox Optimization MATLAB to define, simultaneously, the best position for the wells and best
flow rates for each well in each of the three defined control cycles. During the optimization process was performed
a series of function evaluations using a Reservoir Simulator. Due to the high cost of this process and aiming to
avoid it, a methodology with adaptive surrogate models was employed here. As the optimization problem is
restricted, using an adaptive penalty method allowed the GA to run smoothly. The Egg Model is the study reservoir
in this work. It was executed 20 optimizations to verify the uniformity of the obtained results. The best solution
improved the NVP by 45.32% as compared with the original case. The methodology suggested here brought
consistent results with significant improvements in the NPV, the main objective of this paper.
