Improved Numerical Modeling of Electro-Thermal Coupling of MOVs Based on PSO Algorithm

Metal oxide varistors (MOVs) have been widely employed in DC interruptions to absorb system energy and clamp overvoltage. Specifically, for operating conditions that require continuous actions, such as reclosing, the cumulative effect of temperature may affect the electrical characteristics, which in turn impacts the equalization performance of the whole component. A numerical simulation model that considers the electrical-thermal coupling effect is proposed in this paper. The proposed model employs the particle swarm optimization (PSO) algorithm for parameter design, demonstrating high accuracy and strong robustness. Then, the process of electrical-thermal coupling calculation is introduced. The improved model exhibits better fitting performance and lower energy errors when compared to model without temperature corrections. To validate the accuracy of the model, a MOV temperature testing platform is established. In terms of electrical characteristic predictions, the model achieves a maximum mean absolute percentage error (MAPE) of 0.505% and a maximum energy error of 0.451% across various temperatures, both of which are below 1%. When it comes to temperature predictions, the absolute prediction errors for all varistors remain below 2
$^{\circ }$
C. Comparative analysis of experimental and simulation results indicates that the model is effective in predicting the electro-thermal properties and energy absorption of MOVs.