Insulated Gate Bipolar Transistor Junction Temperature Estimation Technology for Traction Inverters Using a Thermal Model DOI Open Access

Kijung Kong,

Junhwan Choi,

Geonhyeong Park

et al.

Electronics, Journal Year: 2025, Volume and Issue: 14(5), P. 999 - 999

Published: March 1, 2025

This study proposes a method for estimating the junction temperature of power semiconductors, particularly IGBTs (Insulated Gate Bipolar Transistors) and diodes. Traditional measurement methods using NTC (Negative Temperature Coefficient) sensors have limitations in reflecting dynamic conditions real time, as changes take time to reach sensors. To address this, this estimation RC curve fitting thermal impedance model. model represents behavior diodes Foster network that considers resistance capacitance heat transfer path. In particular, transient considering coupling enables prediction verify proposed method, experiments were conducted build based on data measured with an infrared camera The model’s estimated results compared actual values across 25 operating regions, achieving maximum MAE (Mean Absolute Error) 2.26 °C. A comparative analysis first-, second-, third-, fourth-order networks revealed that, while higher orders improve accuracy, gains beyond second order are minimal relative computational demands. contributes enhancing not only reliability semiconductor modules but also minimizing margin inverters by better performance than achieved

Language: Английский

Insulated Gate Bipolar Transistor Junction Temperature Estimation Technology for Traction Inverters Using a Thermal Model DOI Open Access

Kijung Kong,

Junhwan Choi,

Geonhyeong Park

et al.

Electronics, Journal Year: 2025, Volume and Issue: 14(5), P. 999 - 999

Published: March 1, 2025

This study proposes a method for estimating the junction temperature of power semiconductors, particularly IGBTs (Insulated Gate Bipolar Transistors) and diodes. Traditional measurement methods using NTC (Negative Temperature Coefficient) sensors have limitations in reflecting dynamic conditions real time, as changes take time to reach sensors. To address this, this estimation RC curve fitting thermal impedance model. model represents behavior diodes Foster network that considers resistance capacitance heat transfer path. In particular, transient considering coupling enables prediction verify proposed method, experiments were conducted build based on data measured with an infrared camera The model’s estimated results compared actual values across 25 operating regions, achieving maximum MAE (Mean Absolute Error) 2.26 °C. A comparative analysis first-, second-, third-, fourth-order networks revealed that, while higher orders improve accuracy, gains beyond second order are minimal relative computational demands. contributes enhancing not only reliability semiconductor modules but also minimizing margin inverters by better performance than achieved

Language: Английский

Citations

0