<p>This study focuses on the accurate estimation of the inductor core losses, which has become critical with the increasing frequency and power density in power electronics applications such as DC battery charging. The core loss estimation is challenging for powder material core inductors used in medium-frequency interleaved quadratic boost converters due to their complex and nonlinear magnetic behaviour. To overcome this challenge, the parametric simulation analyses were performed within the 20–40&#xa0;kHz frequency range, under varying inductor currents (6–20&#xa0;A) and number of turns (50–80 turns) using the ANSYS-Electronics parametric setup. The obtained data showed that the inductor current has a greater impact on core losses than frequency and number of turns. Using this parametric data, an empirical formula for predicting core losses (avgCoreLoss) was derived using the Genetic Programming (GP) algorithm. The model’s accuracy was verified with Mean Absolute Percentage Error (MAPE) values of 11.50% and R² of 0.9874 for the training data, and MAPE values of 10.37% and R² of 0.9851 for the test data. The developed GP-based model simplifies core loss estimation and provides reliable results with the high accuracy, reducing the need for time-consuming simulations and the more expensive experimental setups.</p>