Load Capacity Dynamic Assessment Method for Fully-sealed Converter Under Random Convection

It is difficult to accurately assess the thermal safety of power devices in fully-sealed converter due to the random convection, which seriously affects the validity of the load capacity evaluation of the converter. Therefore, this paper proposed a load capacity dynamic assessment method of power devices for fully-sealed converters under random convection. Firstly, the random sample of convective thermal resistance of a converter shell was extracted, based on historical operation data and thermal network model. Next, the random simulation of the convective thermal resistance of the shell was realized by using wavelet packet transform and Markov chain principle. Then, the junction temperature fluctuation of the power device was simulated, based on the sequence of convective thermal resistance and load current levels. Finally, the optimal load current limit of the converter was determined by the over-temperature risk of the power devices and growth function. An experimental platform for H-bridge converter with controllable environmental convection was designed. The relevant experimental results show that the proposed method can effectively improve the reliability and economy of the fully-sealed converter.