Real-Time Internal Temperature Estimation and Health Monitoring for IGBT Modules

Authors

Ze Wang, University of Nebraska - LincolnFollow

First Advisor

Wei Qiao

Second Advisor

Jerry L. Hudgins

Third Advisor

George Gogos

Date of this Version

Spring 1-2017

Document Type

Article

Citation

Ze Wang, “Real-time internal temperature estimation and health monitoring for IGBT modules,” Ph.D. dissertation, Dept. Elect. & Comp. Eng., Univ. of Nebraska, Lincoln, 2017.

Comments

A DISSERTATION Presented to the Faculty of The Graduate College at the University of Nebraska In Partial Fulfillment of the Requirements For the Degree of Doctor of Philosophy, Major: Electrical Engineering, Under the Supervision of Professor Wei Qiao. Lincoln, Nebraska: January, 2017

Copyright (c) 2017 Ze Wang.

Abstract

Field experiences have demonstrated that power semiconductor devices, such as insulated-gate bipolar transistors (IGBTs), are among the most fragile components of power electronic converters. Thermomechanical stresses produced by temperature variations during operational and environmental loads are the major causes of IGBT degradation. As the devices are often operated under complex working conditions, temperature variations and the associated damage are difficult to predict during the converter design stage. A promising approach—online health monitoring and prognosis for power semiconductor devices—that can avoid device failure and effectively schedule maintenance has attracted much interest.

This dissertation research focused on real-time accurate internal temperature estimation and health condition monitoring for IGBT modules, where real-time means negligible latency. The objectives of this dissertation research were to: 1) develop low-computational-cost thermal models to accurately estimate important internal temperatures, e.g., junction temperature, in real time for better device overtemperature protection and lifetime prediction; and 2) invent a practical technique to monitor the major aging processes of IGBT modules for effective maintenance scheduling.

To achieve the first objective, this research developed a cost-effective, physics-based thermal equivalent circuit model and a low-order digital filter-based thermal model for IGBT modules. Further, a thermal model adaptive to the solder crack of an IGBT module was developed. This model allows accurate junction temperature monitoring over a device’s lifespan. Additionally, an accurate, frequency domain, transient temperature estimation method for weak points of IGBT modules was developed. The method can be used to determine the bottleneck weak point(s) of IGBT modules. The accurately estimated internal temperatures can be used for more effective operation management of power electronic converters and a more accurate estimation of a device’s remaining useful lifetime (RUL).

To achieve the second objective, a new method of using two-dimensional (2D) case temperatures for real-time monitoring of typical weak points, i.e., bond wire and solder interface, of IGBT modules was invented. This method is based on the physics of failure of IGBT modules and can easily be implemented without complex circuitry. Compared with existing methods, the proposed method is simple to use and does not interrupt the operation of power electronic converters.

Advisor: Wei Qiao