<p>The numerical-physical decomposition concept of hybrid testing provides a new approach to addressing fluid–structure coupling problems. However, online real-time hybrid testing (RTHT) is limited by facility conditions, resulting in insufficiently accurate hydrodynamic measurements. Fortunately, in recent years, the newly developed offline iterative hybrid testing (OIHT) effectively mitigates the challenges associated with high-precision control systems and time delays in online RTHT. Nonetheless, the convergence of OIHT remains an urgent issue that needs to be addressed. Therefore, this paper presents a method which innovates the signal mapping iteration (SMI) algorithm and a hydrodynamic decomposition and updating (HDU) method in the OIHT. The method involves separate analyses and computations for fluid and structure with numerical calculations alongside servo loading systems throughout the entire seismic event. By ensuring the conservation of variables at the fluid–structure interface, the method iteratively corrects response signals between the numerical and physical substructures to achieve precise coupled computation. Through convergence analysis, it has been found that the iteration algorithm and model decomposition are key factors influencing convergence, which forms the basis for the proposed SMI algorithm and HDU method. Finally, the effectiveness of the method is validated through simulation analysis of three single-degree-of-freedom underwater structures. This approach takes advantage of entire time history iteration, enhancing the accuracy of hybrid testing techniques for hydrodynamic measurements. Furthermore, the application of the SMI algorithm and HDU method further improves the convergence and generalizability of the OIHT, providing insights for research on other similar environmental loads, such as wind and waves.</p>