Research on a New Online Junction Temperature Estimation Method for Power Chips and Its Corresponding Technical Requirements

Abstract In this paper, a temperature sensitive electrical parameter (TSEP) method for insulated gate bipolar transistor (IGBT) modules junction temperature estimation based on Voltage rise time is proposed. Firstly, combined with the principles of thermodynamic physics, the movement characteristics of internal carriers and holes were discussed. Then, the voltage and current characteristics of the device's turn-off process were studied. Secondly, compared with the traditional junction temperature detection methods, a junction temperature estimation method based on the voltage rise time was proposed, including the junction temperature model obtained by direct fitting, the junction temperature model based on the mechanism model, and the junction temperature model based on the mechanism and intelligent algorithm, and its feasibility was demonstrated.Finally, the advantages and disadvantages of the three models and the key technologies they rely on were analyzed. The research results show that the detection results of the junction temperature detection method based on the voltage rise time are relatively accurate because it involves both mechanism analysis and online extraction research simultaneously especially the model based on the mechanism model and intelligent algorithm. More importantly, for the online status detection of chips, not only accurate mechanism research is required, but also strong software and hardware design capabilities are needed to meet the requirements of online detection technology. Further reflection reveals that at this stage, compound researchers (theoretical research + technical research + artificial intelligence technology research) are becoming increasingly important, and higher education must play a significant role in this process.