TY - GEN
T1 - A Data-based IGBT Model for Efficient and Accurate Electro-thermal Analysis
AU - Wang, Jianpeng
AU - Xu, Meng
AU - Zhang, Jin
AU - Wang, Laili
AU - Gan, Yongmei
AU - Yamazaki, Tomoyuki
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/10/11
Y1 - 2020/10/11
N2 - Due to the different time-scale behaviors of electricity and thermal, the most electro-thermal IGBT models cannot achieve a good trade-off between simulation efficiency and accuracy. This paper presents a data-based IGBT model for electro-thermal analysis to attain a better performance in circuit simulation. In the model, the data come from the Hefner model which has been re-expressed by MATLAB script. What's more, a two-layer feed-forward neural network with sigmoid hidden neurons and linear output neurons is created to describe the transient behaviors of IGBT devices. The network is trained with the dataset obtained by modifying different parameters in re-expressed Hefner model. Due to the application of IGBT physical model and neural network, the model can be used to simulate not only the transient electricity behaviors, such as switching energy loss and the voltage overshoot, but also the long-time thermal behavior. The complete model is implemented in Simulink and verified by Saber and the experimental prototype test, which shows high efficiency and accuracy.
AB - Due to the different time-scale behaviors of electricity and thermal, the most electro-thermal IGBT models cannot achieve a good trade-off between simulation efficiency and accuracy. This paper presents a data-based IGBT model for electro-thermal analysis to attain a better performance in circuit simulation. In the model, the data come from the Hefner model which has been re-expressed by MATLAB script. What's more, a two-layer feed-forward neural network with sigmoid hidden neurons and linear output neurons is created to describe the transient behaviors of IGBT devices. The network is trained with the dataset obtained by modifying different parameters in re-expressed Hefner model. Due to the application of IGBT physical model and neural network, the model can be used to simulate not only the transient electricity behaviors, such as switching energy loss and the voltage overshoot, but also the long-time thermal behavior. The complete model is implemented in Simulink and verified by Saber and the experimental prototype test, which shows high efficiency and accuracy.
KW - IGBT model
KW - electro-thermal simulation
KW - junction temperature
KW - neural network
UR - https://www.scopus.com/pages/publications/85097131436
U2 - 10.1109/ECCE44975.2020.9236182
DO - 10.1109/ECCE44975.2020.9236182
M3 - 会议稿件
AN - SCOPUS:85097131436
T3 - ECCE 2020 - IEEE Energy Conversion Congress and Exposition
SP - 3442
EP - 3448
BT - ECCE 2020 - IEEE Energy Conversion Congress and Exposition
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 12th Annual IEEE Energy Conversion Congress and Exposition, ECCE 2020
Y2 - 11 October 2020 through 15 October 2020
ER -