Abstract: During high-speed flight, the significant coupling effect between high dynamic pressure and aerodynamic heating-induced thermal gradients causes substantial cumulative aerothermoelastic deformation, which critically impacts flight performance and safety. Currently, cumulative thermal deformation is typically predicted using multi-physics coupling methods; however, research regarding the influence of the coupling time step, a critical parameter in such analyses, remains relatively scarce. To address this gap, the present study employs a fully numerical aerodynamic-thermal-structural two-way coupling method developed by our research team. Taking a two-dimensional flat plate structure as the research object, we systematically investigate the influence of coupling time step on cumulative thermal deformation predictions under various angles of attack and Mach numbers. The results demonstrate that the coupling time step primarily affects the peak value and the long-time heated steady-state value of the cumulative thermal deformation history: larger time steps lead to greater peak deformation but smaller steady-state deformation. Compared to variations in angle of attack, Mach number exerts a more pronounced influence on time step sensitivity, with higher Mach numbers requiring smaller coupling time steps. When focusing on fine-scale evolution, the time step is set to < 2 s; to balance efficiency and accuracy, a time step of 2 s is adopted (with a maximum deformation deviation of < 12% and a steady-state deviation of < 6%); if only steady-state values are of interest, a time step of 5 s can be used, improving computational efficiency by a factor of about 10. Excessively large time steps may induce non-physical oscillations, necessitating reasonable step size constraints to balance computational efficiency and accuracy. Based on systematic analysis of the relative deformation deviation as a function of coupling time step, Mach number, and angle of attack, a selection strategy for coupling time steps is proposed. These findings provide valuable guidance for efficient and refined prediction of cumulative thermal deformation in long-endurance high-speed flight vehicles.