Metallized film capacitors (MFCs) used in flexible DC transmission converter valves typically operate under harsh high-temperature conditions

making them susceptible to failures such as overheating

expansion

and end-cap detachment. The internal hotspot temperature

as a critical factor influencing the rate of insulation aging

significantly affects the operational lifespan of the capacitor. However

due to the difficulty in accurately acquiring thermal parameters on-site

real-time prediction of the temperature distribution remains a challenge. This paper proposes a multivariable inversion model for the thermal parameters of MFCs based on the whale optimization algorithm (WOA). By formulating an objective function of surface temperature distribution derived from the internal heat transfer process

the model employs an inverse problem-solving approach to reconstruct the internal temperature field

heat generation losses

and anisotropic thermal conductivity of the MFC. The average relative error of the inversion results is only 2%. The validity of the proposed model is further verified through finite element simulations and temperature rise experiments. This study provides a theoretical foundation and methodological reference for engineering applications involving temperature distribution and thermal parameter prediction of MFCs.