Thunderstorm and Cloud Algorithm: A Novel Parameter-Free Metaheuristic Inspired by Atmospheric Dynamics for Complex Optimization Tasks

Abstract

Solving complex, nonlinear, and high-dimensional optimization problems remain a central challenge in engineering, science, and artificial intelligence. Traditional deterministic methods often struggle in such landscapes due to their dependence on gradients and rigid mathematical assumptions. To overcome these limitations, this paper proposes the Thunderstorm and Cloud Algorithm (TCA)—a novel, parameter-free metaheuristic inspired by atmospheric processes observed during thunderstorms. TCA models the behaviors of cloud formation, lightning strikes, wind movement, and rainfall to guide its search strategy. Each mechanism is assigned a specific optimization role: clouds represent candidate solutions, lightning drives exploitation toward the best regions, wind introduces global diversity, and rainfall enables focused local refinement. A key innovation of TCA lies in its parameter-free design, removing the burden of tuning and enhancing robustness across problem domains. The algorithm’s performance was rigorously tested on the 29-function CEC 2017 benchmark suite and compared with nine recent state-of-the-art metaheuristics. Results reveal that TCA ranked first across all benchmark functions, outperforming its competitors in terms of solution quality, convergence behavior, and robustness. Overall, TCA introduces a biologically inspired, adaptive, and highly competitive optimization framework. Its superior performance and simplicity make it a promising tool for solving diverse, real-world optimization problems.