An Improved Electromagnetic Field Optimization for the Global Optimization Problems

Area of Science:

• Computational Intelligence
• Metaheuristic Optimization
• Physics-Inspired Algorithms

Background:

• Electromagnetic Field Optimization (EFO) is a population-based metaheuristic algorithm inspired by electromagnetism and the golden ratio.
• EFO utilizes attraction-repulsion forces among electromagnetic particles to navigate optimization landscapes, balancing global and local search.
• Existing metaheuristic algorithms often face challenges in balancing exploration and exploitation effectively.

Purpose of the Study:

• To introduce an improved version of the Electromagnetic Field Optimization algorithm, termed improved Electromagnetic Field Optimization (iEFO).
• To enhance the performance of EFO by incorporating novel solution generation and adaptive parameter control mechanisms.
• To rigorously evaluate the efficacy of iEFO against established algorithms.

Main Methods:

• Development of iEFO with a new solution generation function for electromagnets.
• Implementation of adaptive control for algorithmic parameters within iEFO.
• Modification of boundary control and randomization procedures for electromagnet generation.

Main Results:

• Computational studies demonstrated that iEFO significantly outperforms the original EFO algorithm.
• iEFO showed superior performance compared to other state-of-the-art metaheuristic algorithms, including Artificial Bee Colony, Particle Swarm Optimization, and Differential Evolution.
• Statistical testing verified the significant improvements achieved by the proposed iEFO.

Conclusions:

• The proposed iEFO algorithm represents a significant advancement in physics-inspired metaheuristic optimization.
• iEFO's novel modifications lead to superior performance in solving complex optimization problems.
• The adaptive nature and improved generation functions of iEFO enhance its ability to find optimal solutions effectively.