Open AccessResearch on Life Prediction of IGBT Devices Based on Elman Neural Network Model
Author(s) -
C.Y. Kong L.J. Xing,
Xinze Xi,
Xiaobin He,
Shengnan Li,
Mingqun Liu,
Lie Li
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1550/3/032148
Subject(s) - insulated gate bipolar transistor , artificial neural network , reliability (semiconductor) , exponential function , arrhenius equation , power (physics) , computer science , voltage , reliability engineering , artificial intelligence , electrical engineering , engineering , activation energy , mathematics , physics , thermodynamics , mathematical analysis , chemistry , organic chemistry
IGBT, as a new generation of composite full-controlled voltage-driven power semiconductor devices, has been widely used in the field of modern power electronics. The study of IGBT device life prediction has important guiding significance for the stable operation and reliability management of power system. In this paper, the Cofin-Manson-Arrhenius extended exponential model based on thermal load is analyzed, and the model parameters are trained and modified by the Elman neural network in order to improve the prediction accuracy of life prediction model. The Coffin-Manson-Arrhenius extension exponential model and the new improved model are simulated and validated by experiments. The results are compared with the actual life of IGBT. It is concluded that the Coffin-Manson-Arrhenius extension exponential model based on Elman neural network has higher accuracy in life prediction than the Coffin-Manson-Arrhenius extension exponential model.