Spiny Fish Optimization Algorithm

To address issues such as premature convergence, dimensionality imbalance convergence, excessive reversibility of search paths, and population structure degradation in complex nonlinear, multimodal, and high-dimensional optimization problems, this paper proposes an irreversible topological bistate stickleback optimization algorithm. Based on an abstract model of stickleback territoriality and breeding-stage attack behavior, the algorithm introduces a bistate dynamics mechanism, an information density-driven territory topology, an anisotropic attack tensor, an adaptive topology contraction mechanism, and an energy-risk coupled field model during population evolution. Simultaneously, an irreversible search path is constructed through a safe zone solidification mechanism, thereby enhancing local convergence stability while ensuring global exploration capabilities. This paper establishes a unified update equation from a dynamical system perspective, analyzes the algorithm's convergence characteristics and stability conditions, and provides a time complexity analysis and theoretical property explanation. This method reconstructs the swarm intelligence search model from the perspective of topological and energy coupling, providing a new theoretical framework for constructing high-dimensional stable optimization algorithms.