<p>Triboelectric nanogenerators (TENGs) are versatile energy-harvesting devices that convert mechanical motion (often low-frequency) into electricity through contact-separation, or sliding modes. They are increasingly explored for renewable energy harvesting in wind-, wave-, and other fluid-driven environments. Effective optimization of TENGs necessitates consideration of mechanical motion, electrostatic charge transfer, and, particularly in fluid-driven systems, complex flow behaviors. Numerical simulation is indispensable for device design; however, most of the existing reviews address these domains in isolation and often neglect the relationship among modeling assumptions, coupling strategies, and predictive accuracy. This review presents a unified framework that integrates structural mechanics, electrostatics, computational fluid dynamics (CFD), and fluid–structure interaction (FSI). As the central concept, <i>coupling depth</i>, is introduced to define the degree of dynamic, bidirectional interaction among physical domains, ranging from single-physics to fully coupled multiphysics simulations. This framework facilitates systematic assessment of model fidelity, computational cost, and reproducibility. The review further examines the impact of modeling choices on simulation outcomes, identifies reproducibility gaps arising from incomplete reporting and implicit assumptions, and proposes a modeling roadmap that emphasizes reduced-order models, machine-learning surrogates, and digital twin technologies as prospective research directions.</p> Graphical Abstract