Application of particle swarm optimization and genetic algorithm for optimization of a southern Iranian oilfield

Abstract

Optimization of the placement and operational conditions of oil wells plays an important role in the development of the oilfields. Several automatic optimization algorithms have been used by different authors in recent years. However, different optimizers give different results depending on the nature of the problem. In the current study, a comparison between the genetic algorithm and particle swarm optimization algorithms was made to optimize the operational conditions of the injection and production wells and also to optimize the location of the injection wells in a southern Iranian oilfield. The current study was carried out with the principal purpose of evaluating and comparing the performance of the two most used optimization algorithms for field development optimization on real-field data. Also, a comparison was made between the results of sequential and simultaneous optimization of the decision variables. Net present value of the project was used as the objective function, and the two algorithms were compared in terms of the profitability incremental added to the project over twelve years. First, the production rate of the producers was optimized, and then water alternating gas injection wells were added to the field at locations determined by engineering judgment. Afterward, the location, injection rate, and water alternating gas ratio of the injectors were optimized sequentially using the two algorithms. Next, the production rate of the producers was optimized again. Finally, a simultaneous optimization was done in two manners to evaluate its effect on the optimization results: simultaneous optimization of the last two steps and simultaneous optimization of all decision variables. Results showed the positive effect of the algorithms on the profitability of the project and superiority of the particle swarm optimization over the genetic algorithm at every stage. Also, simultaneous optimization was beneficial at finiding better results compared to sequential optimization approach. In the end, a sensitivity analysis was made to specify the most influencing decision variable on the project’s profitability.

Hybrid Particle Swarm Optimization Coupled by Machine Learning Technique for Estimating Relative Permeability and Capillary Pressure Functions During Low-Salinity Flood

“…Various optimization methods have been adopted in the AHM process, including, Gradient-Based, Hybrid, Stochastic (Genetic Algorithm (GA), Particle Swarm Optimization (PSO)), and Probabilistic methods. Each method has positives and negatives [4,5,[8][9][10][11][12][13]. Finding the best HM algorithm among several available methods, is the objective of many conducted studies.…”

Low salinity water flooding: estimating relative permeability and capillary pressure using coupling of particle swarm optimization and machine learning technique

“…Based on the purpose of our study, in which HM is conducted at a core scale (ENKF is not necessary) and running time is important, PSO is preferred. The reason is that the optimized parameters will be determined faster and less scattered than GA 10 , 12 .…”