Genetic Algorithm Assisted Fuzzy Iterative Learning Optimizer for Automatic Optimization of Oil well Placement under Production Constraints

In this paper a new optimization approach based on fuzzy systems and iterative learning is proposed where Genetic Algorithm (GA) employed to optimally determine fuzzy parameters. The method is appropriate for highly nonlinear and uncertain large scale systems such as optimal oil well placement. Well-placement is a crucial step in field development. However, the major difficulties of the problem are highly nonlinear dynamics of reservoir, well locations constraints and large number of decision variables. Therefore, in this paper, a new optimization method is proposed and employed to solve the problem. Fuzzy rule generation is done employing GA to avoid being stuck in local optima. Since fuzzy coefficients are considered as decision variables instead of well locations, number of optimization parameters reduces significantly. Simulation results show superior performance such as lower computational load and less number of simulator runs compared with ones obtained by previous methods.