Superconducting magnets in particle accelerators are subject to various forces, radiation, high voltages, or thermal gradients. These can cause failures such as inter-turn shorts, which can be very challenging to detect after the magnets are installed in the machine. Measurements in the time- and frequency domain can pinpoint outlier magnets or precursors of such failures in the data. Interpretable, physics-driven simulations can facilitate this by explaining if manufacturing tolerances cause the observed behaviour or if it can be correctly classified as an outlier. An equivalent circuit model is proposed to simulate superconducting accelerator magnets’ impedances. The physics-driven model includes various coupled, non-linear effects in the superconductors, such as superconducting filament magnetisation, inter-filament and inter-strand coupling currents, eddy currents in the strands’ copper and various magnet components, and stray capacitances. The model is validated
against a wide range of available time- and frequency-domain measurements. A very good agreement is achieved across the different measurements and domains. The network model is computationally very inexpensive while still preserving good accuracy and interpretability. The model can hence be used to reproduce impedances of superconducting magnets accurately for performance evaluations and to investigate the impact of failures.