Drosophila Optimization Algorithm Based on Chaotic Development Mechanism and Orthogonal Learning Strategy for...

Area of Science:

• Petroleum Engineering
• Computational Intelligence
• Optimization Algorithms

Background:

• Subsurface production optimization is critical for economic sustainability in the oil and gas industry.
• Conventional methods face challenges with high computational costs and limited effectiveness.
• Evolutionary algorithms offer a promising gradient-free approach for complex optimization tasks.

Purpose of the Study:

• To introduce a novel algorithm, the chaotic exploitation orthogonal learning fruit fly optimization algorithm (COFOA), for global and oil/gas production optimization.
• To enhance the balance between exploration and exploitation in optimization processes.
• To improve the efficiency and effectiveness of reservoir management decisions.

Main Methods:

• Integration of a chaotic exploitation mechanism to escape local optima and improve search efficiency.
• Incorporation of an orthogonal learning strategy to strengthen the algorithm's exploitation capability.
• Extensive testing on benchmark functions (IEEE CEC 2017, 2022) and real-world reservoir production optimization scenarios.

Main Results:

• COFOA demonstrated significant performance superiority over existing algorithms in reservoir production optimization.
• Achieved net present value (NPV) improvements ranging from 2.35% to 16.23% compared to state-of-the-art methods.
• Outperformed competitors like mSCA, BLPSO, SCADE, CCMSCSA, HGWO, and CCMWOA in terms of mean NPV.

Conclusions:

• The proposed COFOA algorithm effectively addresses the limitations of conventional reservoir optimization techniques.
• COFOA exhibits superior global optimization capabilities, particularly for maximizing NPV in complex reservoir conditions.
• The integration of chaotic exploitation and orthogonal learning enhances search efficiency and exploitation effectiveness.