Real-Time Average Junction Temperature Estimation for Multichip IGBT Modules With Low Computational Cost

The junction temperature of the insulated gate bipolar transistor (IGBT) modules is a vital parameter for the reliability of the power electronic system. For effective thermal management, it is necessary to estimate junction temperature in real time. However, existing thermal models have limitations to achieve fast temperature estimation for multichip IGBT modules. In this article, we investigate the real-time junction temperature estimation for multichip IGBT modules with low computational cost. According to the power module' structure, an average 2-D thermal model with physical significance is proposed to estimate the virtual junction temperature for multichip IGBT modules, which can be extracted from finite element model based on the least square method. To further improve the computational efficiency, a temperature estimator based on the staggered cycle-by-cycle calculation method is proposed and its estimation error under various extreme operation conditions is analyzed by enumeration method. Finally, the comprehensive experimental results verify the effectiveness of the proposed average thermal model and junction temperature estimation method.