Optimization of Injection-Production Methods Based on Improved Genetic Algorithm and Automated Simulation with ECLIPSE

Abstract The optimization of injection-production methods plays a crucial role in enhancing the efficiency and economic benefits of oil reservoir development. Traditional adjustments of injection-production parameters and methods primarily rely on manual tuning, which is cumbersome, time-consuming, and prone to falling into local optima. This makes it difficult to systematically evaluate the advantages and disadvantages of various injection-production modes, such as separate injection with cyclic production, cyclic injection with cyclic production, cyclic injection with separate production, and separate injection with separate production. Such an approach not only limits the optimization space of development schemes but may also lead to resource wastage and reduced development effectiveness. To address this issue, this paper proposes an automatic optimization method for injection-production strategies based on an improved genetic algorithm. Building upon the traditional genetic algorithm, adaptive mutation strategies and elite preservation mechanisms are integrated to significantly enhance the algorithm's global search capability and convergence speed. The adaptive mutation strategy dynamically adjusts the mutation probability according to the evolutionary state of the population, improving the algorithm's ability to escape local optima. The elite preservation mechanism ensures the inheritance of superior individuals, allowing the algorithm to retain optimal solutions during iterations. By developing automated scripts, a seamless integration between the improved genetic algorithm and the ECLIPSE reservoir numerical simulation software is achieved, constructing a fully automated optimization platform for injection-production methods. This platform can automatically generate different combinations of injection-production methods, including modes such as separate injection with cyclic production, cyclic injection with cyclic production, cyclic injection with separate production, and separate injection with separate production. It invokes ECLIPSE for numerical simulation and quantitatively evaluates each scheme based on optimization objective functions such as cumulative oil production, recovery factor, water cut, and economic indicators. Through iterative optimization, the algorithm can automatically identify and output the optimal injection-production method without manual intervention. This method achieves full automation in the optimization of injection-production strategies, enhances work efficiency, and reduces the impact of human factors on the optimization results.