Dynamic gold rush optimizer: fusing worker adaptation and salp navigation mechanism for enhanced search

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Metaheuristics

Background:

• The Gold Rush Optimizer (GRO) faces challenges with premature convergence and suboptimal solutions.
• Existing optimization methods often struggle with balancing exploration and exploitation effectively.

Purpose of the Study:

• To introduce the Dynamic Gold Rush Optimizer (DGRO), an advanced variant of GRO.
• To address the limitations of GRO by enhancing its exploration and exploitation capabilities.
• To improve convergence towards global optima in complex optimization problems.

Main Methods:

• Developed DGRO by integrating the Salp Navigation Mechanism (SNM) and Worker Adaptation Mechanism (WAM).
• SNM provides dynamic population guidance for effective solution space navigation and smooth exploration-exploitation transition.
• WAM enhances exploration through localized interactions, promoting adaptive learning in promising search regions.

Main Results:

• DGRO demonstrated superior performance and stability across CEC2013 and CEC2020 benchmark functions in 30 and 50-dimensional spaces.
• Experimental results on seven complex engineering optimization problems confirmed DGRO's effectiveness.
• Statistical analyses (WRST, FRT) validated the significant advancements of DGRO over existing methods.

Conclusions:

• DGRO effectively overcomes the limitations of GRO, particularly premature convergence.
• The novel SNM and WAM mechanisms significantly enhance optimization capability and robustness.
• DGRO represents a competitive and stable advancement in the field of optimization algorithms.