Abstract: With the large-scale application of HVDC transmission technology, the development and verification of control strategies for power electronic devices increasingly rely on high-precision real-time simulation platforms, and the modeling accuracy of transmission lines directly affects the reliability of simulation results. As the industry-recognized most accurate modeling method for HVDC transmission lines, the traditional phase-domain frequency-dependent model usually adopts linear interpolation to reconstruct historical current data when dealing with non-integer time delays. However, during the transient process of power electronic devices, the linear interpolation method is difficult to accurately reflect the nonlinear variation of current. Additionally, there are differences in the transient responses of different HVDC transmission lines, making it hard to correct interpolation deviations through fixed formulas, which limits the simulation accuracy. To address this issue, a dynamic adaptive interpolation method based on artificial intelligence is proposed. By optimizing interpolation parameters through data-driven approaches, the simulation accuracy of transient processes is effectively improved. Experimental results show that this method can significantly reduce the errors introduced by traditional interpolation methods, providing more accurate simulation support for the stability analysis of HVDC transmission systems and the optimization of control strategies.