Optimization as Ecological Survival: A Topological Humidity–Based Niche Selection Framework

To address the common reliance on "forced position updates" and "global optimum guidance" in existing swarm intelligence optimization algorithms, this paper proposes a novel swarm intelligence optimization algorithm—the American Green Tree Frog Optimization Algorithm. This algorithm no longer treats individual position updates as the default behavior, but instead introduces "existence probability" as a core state variable, remodeling the optimization process as an ecosystem evolution process within the solution space. Within this framework, whether an individual continues to exist, whether it moves, and how it searches are all determined by its ecological value, adaptive stability, and local environmental structure. The algorithm perceives the solution space structure through a topological humidity field, maintains diversity through niche occupation and competition mechanisms, and avoids premature convergence through self-destruction and structural regeneration mechanisms. Theoretical analysis shows that this algorithm differs from existing swarm intelligence algorithms in its search mechanism, decision logic, and convergence form, exhibiting excellent generalization ability and structure perception capability.