Application of IGBT junction temperature estimation in hybrid electric vehicles

The Insulated Gate Bipolar Transistor (IGBT) is a crucial component in hybrid electric vehicle electric drive systems. Estimating the IGBT junction temperature is essential for forecasting its lifespan and managing the heat within the power module. This paper presents and applies two methods for predicting IGBT junction temperature: the thermal network model and the empirical relationship method. Among them, the thermal impedance method is analyzed and extended to engine application by introducing a coefficient beta = f (speed, coolant boundary). The empirical relationship is first proposed considering the effect of thermal boundary. Both methods are applied in CHANGAN Plug-in Hybrid Electric Vehicles (PHEV). Experimental results indicate that the thermal network method provides fast and accurate temperature estimates with the absolute error remaining below +/- 5 % under over 99 % operating conditions. In contrast, the empirical relationship method offers lower frequency but solid predictions under separate working conditions. Both methods were evaluated through road tests and demonstrated the capability to provide reasonable temperature variations of the power module.