Application of an improved harmony search algorithm in well placement optimization using streamline simulation

Optimal well placement is a crucial step in efficient reservoir development process which significantly affects the productivity and economical benefits of an oil reservoir. However, it is a complex and challenging problem due to the different engineering, geological and economical variables involved. This leads to a very large number of potential scenarios that must be evaluated using numerical reservoir simulations. The key points in such an optimization process are using a fast function evaluation tool and development of an efficient and robust optimization algorithm that can find good solutions with a minimum required number of function evaluations. This study presents an approach that employs an Improved Harmony Search (IHS) algorithm in combination with streamline simulation to determine the optimum well locations in a reservoir. Several case studies including both synthetic and real ones have been considered, and the performance of IHS has been compared to that of some stochastic optimization algorithms including Particle Swarm Optimization (PSO), Simulated Annealing (SA), Genetic Algorithm (GA), and the classical Harmony Search (HS) algorithm. Results of the case studies demonstrate the superiority of IHS algorithm over the other mentioned algorithms in providing fast and accurate solutions. These findings are very promising as the IHS is a very simple and easy-to-implement algorithm which can be widely used in other areas of reservoir development process.

► We compared the performance of the IHS algorithm to that of the HS, PSO, SA and GA in well placement optimization problems.
► A streamline based simulator has been used as the objective function evaluator to speed up the optimization procedure.
► These algorithms have been compared in solving well placement optimization problem of four study cases.
► In all of the cases, IHS algorithm outperformed other algorithms and produced better solutions.