An enhanced fruit fly optimization algorithm with random spare and double adaptive weight strategies for oil and gas production optimization

In the field of petroleum extraction, enhancing oil and gas recovery processes is essential for sustaining the economic viability of energy enterprises and addressing the continuously increasing global energy demand. Efficient subsurface production plays a pivotal role in strategic decision-making, including the selection of optimal drilling sites and the determination of effective well control parameters. However, conventional reservoir optimization techniques are often computationally intensive and may struggle to deliver satisfactory solutions. As a promising alternative, evolutionary algorithms-rooted in the principles of natural selection-have demonstrated strong potential for addressing complex optimization problems due to their gradient-free nature and inherent suitability for parallel computation. In this study, we propose an enhanced evolutionary algorithm tailored for global optimization and oil and gas production improvement. This method builds upon the original Fruit Fly Optimization Algorithm (FOA) by incorporating a random spare mechanism and a dual adaptive weighting scheme, aiming to achieve a more effective balance between exploration and exploitation during the search process. Specifically, after the standard FOA updates the population, the random spare mechanism is introduced to enhance exploratory capabilities and avoid premature convergence. Subsequently, the dual adaptive weighting strategy is employed to improve convergence speed and solution refinement. The proposed RDFOA algorithm is rigorously validated through comprehensive experiments on benchmark test suites from IEEE CEC 2017 and IEEE CEC 2022. These evaluations include ablation studies, scalability analyses, visualization of search trajectories, and comparative assessments against state-of-the-art algorithms. On the CEC 2017 benchmark, RDFOA outperforms CLACO in 17 functions and QCSCA in 19 functions. On the CEC 2022 benchmark, it surpasses CCMSCSA and HGWO in 10 functions. The experimental results clearly demonstrate that RDFOA consistently achieves superior performance in oil and gas production optimization scenarios.