Improved intelligent clonal optimizer based on adaptive parameter strategy.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Evolutionary Computation

Background:

• The intelligent clonal optimizer (ICO) is a novel evolutionary algorithm utilizing a unique cloning and selection mechanism.
• Existing ICO performance can be limited by convergence speed and population diversity, potentially leading to stagnation.

Purpose of the Study:

• To enhance the performance of the intelligent clonal optimizer (ICO).
• To improve convergence speed, population diversity, and solution accuracy.
• To address optimal value update stagnation in evolutionary algorithms.

Main Methods:

• Integration of quasi-opposition-based and quasi-reflection-based learning strategies to manage exploration-exploitation transitions.
• Implementation of an adaptive parameter method to dynamically adjust algorithm parameters during optimization.
• Development of the improved intelligent chaotic clonal optimizer (IICO) incorporating these enhancements.

Main Results:

• The IICO demonstrated competitive performance across 27 benchmark functions, 8 CEC 2014 test functions, and 3 engineering problems.
• Comparative analysis against ten meta-heuristic algorithms showed superior convergence rate and accuracy.
• The adaptive parameter strategy effectively prevented optimal value update stagnation.

Conclusions:

• The proposed IICO significantly improves upon the ICO by enhancing convergence speed and solution accuracy.
• The integration of quasi-opposition learning and adaptive parameters provides a robust approach to complex optimization.
• IICO exhibits competitive and effective performance in numerical optimization tasks.