This paper presents an approach to combine multiple temperature-sensitive electrical parameters to improve the accuracy and precision of the junction temperature estimation of power transistors using the example of a silicon-carbide power MOSFET. Switching delays and the threshold voltage of the power transistor during turn-on and -off of a silicon-carbide power transistor are used as temperature-sensitive electrical parameters for the online junction temperature measurements. In order to improve the accuracy, a shallow fully-connected neural network is used as the means to combine the four measurements in one switching cycle of the transistor. The maximum measurement error of the junction temperature of the power transistor is reduced approximately 10-fold from 8.98 K to 0.92 K.
%0 Generic
%1 1sharma2022characterization
%A Sharma, Kanuj
%A Kamm, Simon
%A Baron, Kevin Munoz
%A Kallfass, Ingmar
%B 24th European Conference on Power Electronics and Applications (EPE'22 ECCE Europe), Hannover, Germany, September, 2022
%D 2022
%I IEEE
%K 2022ias ias
%T Characterization of Online Junction Temperature of the SiC power MOSFET by Combination of Four TSEPs using Neural Network
%U https://ieeexplore.ieee.org/abstract/document/9907755
%X This paper presents an approach to combine multiple temperature-sensitive electrical parameters to improve the accuracy and precision of the junction temperature estimation of power transistors using the example of a silicon-carbide power MOSFET. Switching delays and the threshold voltage of the power transistor during turn-on and -off of a silicon-carbide power transistor are used as temperature-sensitive electrical parameters for the online junction temperature measurements. In order to improve the accuracy, a shallow fully-connected neural network is used as the means to combine the four measurements in one switching cycle of the transistor. The maximum measurement error of the junction temperature of the power transistor is reduced approximately 10-fold from 8.98 K to 0.92 K.
@conference{1sharma2022characterization,
abstract = {This paper presents an approach to combine multiple temperature-sensitive electrical parameters to improve the accuracy and precision of the junction temperature estimation of power transistors using the example of a silicon-carbide power MOSFET. Switching delays and the threshold voltage of the power transistor during turn-on and -off of a silicon-carbide power transistor are used as temperature-sensitive electrical parameters for the online junction temperature measurements. In order to improve the accuracy, a shallow fully-connected neural network is used as the means to combine the four measurements in one switching cycle of the transistor. The maximum measurement error of the junction temperature of the power transistor is reduced approximately 10-fold from 8.98 K to 0.92 K.},
added-at = {2023-03-13T08:18:56.000+0100},
author = {Sharma, Kanuj and Kamm, Simon and Baron, Kevin Munoz and Kallfass, Ingmar},
biburl = {https://puma.ub.uni-stuttgart.de/bibtex/249c0d196eccba2a290a50a0cbba8c959/taylansngerli},
booktitle = {24th European Conference on Power Electronics and Applications (EPE'22 ECCE Europe), Hannover, Germany, September, 2022},
eventdate = {05.-09.09.2022},
interhash = {966dc1bd97b3f4a53f656638c23cf2c1},
intrahash = {49c0d196eccba2a290a50a0cbba8c959},
keywords = {2022ias ias},
publisher = {IEEE},
timestamp = {2023-03-13T09:27:53.000+0100},
title = {Characterization of Online Junction Temperature of the SiC power MOSFET by Combination of Four TSEPs using Neural Network},
url = {https://ieeexplore.ieee.org/abstract/document/9907755},
venue = {Hannover, Germany},
year = 2022
}
