Energy valley optimizer: a novel metaheuristic algorithm for global and engineering optimization.

Area of Science:

• Computational intelligence
• Optimization algorithms
• Metaheuristic computing

Background:

• Metaheuristic algorithms are crucial for solving complex optimization problems.
• Existing algorithms face challenges with high-dimensional and real-world applications.
• Novel approaches are needed to enhance optimization efficiency and effectiveness.

Purpose of the Study:

• To introduce the Energy Valley Optimizer (EVO), a novel metaheuristic algorithm.
• To evaluate EVO's performance on a suite of unconstrained mathematical test functions across various dimensions.
• To compare EVO against state-of-the-art algorithms using rigorous statistical analysis and competition benchmarks.

Main Methods:

• The Energy Valley Optimizer (EVO) algorithm is developed, inspired by advanced physics principles of particle decay and stability.
• Performance evaluation involves 20 unconstrained mathematical test functions in multiple dimensions.
• Statistical validation includes 100 independent runs, mean, standard deviation, objective function evaluations, and comparative analyses (Kolmogorov-Smirnov, Wilcoxon, Kruskal-Wallis).
• Comparison with leading algorithms from recent Competitions on Evolutionary Computation (CEC) for real-world optimization.

Main Results:

• EVO demonstrates competitive and outstanding performance on complex benchmark functions.
• The algorithm shows efficacy in handling high-dimensional optimization problems.
• Statistical analyses confirm the robustness and reliability of the EVO algorithm.

Conclusions:

• The proposed Energy Valley Optimizer (EVO) is a promising novel metaheuristic algorithm.
• EVO provides effective solutions for complex mathematical benchmarks and real-world optimization challenges.
• The physics-inspired approach offers a competitive alternative to existing state-of-the-art optimization techniques.