In recent years, the demand for high power density power supplies for accelerator systems has been growing rapidly, and the design of high-power modular power supplies is facing serious challenges [1–3]. The EMC design of high power density switching power supplies faces the serious challenge of accurate prediction and suppression of high-frequency conducted disturbances. In this paper, we propose an analysis method that incorporates circuit behavior modeling in concert with electromagnetic interference (EMI) spectral algorithms. Firstly, a cross-domain correlation model between time-domain current waveforms and frequency-domain interference features is established, and the switching transient process is mapped to the conduction path via parasitic parameters through circuit behavior modeling to improve the EMI prediction accuracy in the 150 kHz–30 MHz band. Second, the RBW adaptive FFT algorithm is used, and the EMI curves are compared with the simulation curves, and the experiments show that the method can accurately identify the EMI spectrum, and the spike bands are matched. This study provides a theoretical framework and a practical tool for the EMC design of high-density power modules.