Silver Pomfret Optimization Algorithm

To address the issues of premature convergence, directional degradation, and anisotropic imbalance in high-dimensional complex optimization problems, swarm intelligence algorithms are prone to fall into problems such as premature convergence, directional degradation, and anisotropic imbalance in high dimensions. This paper proposes a silver pomfret optimization algorithm. The algorithm starts from the perspective of swarm fluid dynamics, constructing a unified discrete dynamic system model coupling position, momentum, and phase variables. An anisotropic fluid tensor is introduced to characterize the swarm covariance structure, achieving adaptive manifold compression search. The energy-phase coupling mechanism regulates the swarm oscillation rhythm; a closed-loop regulation system is constructed using the swarm entropy control equation for dynamic equilibrium exploration and development; and an adaptive Lévy burst operator is combined to enhance global escape capability. Theoretically, this paper presents a semi-convergence analysis framework based on candidate Lyapunov functions, proving that the system has a stable contraction trend under certain parameter conditions. The algorithm elevates from heuristic rules to a structured coupled dynamic system, providing a new modeling paradigm for swarm intelligence optimization.