Abstract: To quantify the influence laws of different frequency components on the dynamic response of cable-stayed Bridges, this study takes a certain cross-sea single-tower cable-stayed bridge as the research object, and uses the original ground motion near the fault and the pulse and high-frequency components decomposed by wavelet transform as the ground motion input. The influence of pulse components and high-frequency components on the dynamic response of cable-stayed Bridges is discussed.The interactive coupling of pulse-frequency and stochastic high-frequency components in near-fault ground motions leads to significant uncertainties in the dynamic response of cable-stayed bridges, elevating their potential risk of damage. To quantify the influence patterns of different frequency components on the dynamic response of cable-stayed bridges, it takes a single-pylon sea-spanning cable-stayed bridge as the research object, and uses the original ground motion near the fault and the pulse and residential frequency components decomposed by wavelet transform as the earthquake inputs. The effect of pulse and residential components on the dynamic response of cable-stayed bridge is explored. Based on dimensionless indices characterizing near-fault ground motions and cable-stayed bridge response characteristics, the range of influence of different frequency components on the dynamic response of cable-stayed bridges is revealed, and a predictive model for the dynamic response of cable-stayed bridges considering pulse effects is developed. The study results show that residential frequency has a significant effect on dynamic response of cable-stayed bridge when the pulse period deviates from the bridge’s fundamental period; and that the fitted curves of dimensionless indicators (Φ_VSI, Φ_EPV, Φ_HI, T₁/T_p) and the response ratios of pulse and high-frequency components exhibit an intersecting characteristic. When the values of dimensionless parameters exceed the intersection point, the pulse component becomes the dominant factor in the dynamic response of cable-stayed bridges; and the established model demonstrates predictive errors of 17.8%, 19.2%, and 12.3% for displacements at critical locations including the pylon top, girder end, and pier top respectively, reproducing with reasonable accuracy the nonlinear dynamic response of cable-stayed bridges under near-fault pulse-like ground motions. The methodology presented in this study can serve as a reference for the aseismic design of cable-stayed bridges in near-fault regions.