Research On Junction Temperature Measurement And Prediction Of IGBT Module

Insulated gate bipolar transistor(IGBT)is a composite fully controlled power semiconductor formed by the combination of bipolar junction transistor(BJT)and metal oxide semiconductor field effect transistor(MOSFET).It combines the advantages of MOSFET and BJT and has become an excellent switching device in the field of power electronics.The junction temperature is one of the important indicators that affect the reliability of the device.Effective measurement and prediction of the junction temperature of the IGBT is to ensure the reliable operation of the power device One of the important measures of the IGBT is that the IGBT is subject to long-term continuous thermal shock.The junction temperature fluctuates too high,which causes the device aging failure to become a key factor restricting reliability.Therefore,the measurement and prediction of the IGBT junction temperature during the aging process is reliable for the device Operation is of great significance.This article takes IGBT as the object to study a method that can accurately predict the junction temperature of IGBT module.First,we briefly analyze the basic structure and working characteristics of IGBT,summarize and analyze the basic heat transfer mechanism of IGBT,and understand that frequent fluctuations of temperature and junction temperature are important factors that affect the reliability of IGBT modules.Under the premise of comparing and analyzing different types of junction temperature measurement methods,comprehensive consideration of the correlation between module aging and various parameters,combined with the accuracy of junction temperature measurement,finally selects the saturation voltage drop collector current as the junction temperature measurement of the temperature-sensitive parameter method.Secondly,the actual operating conditions of IGBT are more complicated,and it is difficult to predict the junction temperature under actual operating conditions.In order to simulate the normal operating state of IGBT devices more closely,this paper designs and builds an aging experiment platform,and introduces related experimental methods and steps.Obtain the collector current,saturation voltage drop and junction temperature data under different conditions to serve the subsequent junction temperature prediction.Finally,in order to predict the junction temperature of the IGBT,we built a mathematical analysis model,but found that the precise mathematical analysis model is difficult to achieve,so in order to avoid this tedious work physical mechanism,we combined all the experimental platforms to establish a The IGBT neural network junction temperature prediction model based on the error back propagation algorithm(back propagation,BP)optimized by the simulated annealing algorithm is compared with the mathematical model and the BP neural network model.It is found that the junction temperature prediction model based on experimental data is easier to establish and It is more convenient to use and has higher prediction accuracy.