For optimization problems that are high-dimensional, strongly non-convex, have multiple local optima, and are costly to evaluate the objective function, traditional swarm intelligence optimization methods based on time-step progression generally suffer from large computational redundancy, premature convergence, and a sharp decline in search efficiency with increasing dimensionality. To address this, this paper proposes a novel optimization paradigm: an energy-constrained geometric search method based on an event-triggered mechanism. This method does not rely on fixed time steps for evolution; instead, it uses the occurrence of "captureable opportunities" as the condition for state transitions. Through explicit energy modeling, event-triggered updates, and a geometrically constrained search space, it achieves a sparse, discontinuous, but stable optimization process. Inspired by the ambush and predation mechanism of the white-lipped pit viper in biology, this method does not rely on direct analogies of biological behavior but abstracts it as a geometric optimization problem within a hybrid event system. By introducing event-triggered operators, a reversible freezing mechanism, capture polyhedral constraints, and energy-driven burst update rules, this method exhibits significantly different properties from traditional swarm intelligence algorithms in terms of search efficiency, stability, and multi-modal preservation capabilities. Theoretical analysis shows that this method has bounded update properties and a stable frozen subspace structure under energy constraints, providing a theoretical basis for its application in expensive optimization problems.