TY - GEN

T1 - Inverse thermal model of temperature-to-power mapping for GaN systems

AU - Zhang, Shuangfeng

AU - Laboure, Eric

AU - Labrousse, Denis

AU - Lefebvre, Stéphane

PY - 2018/1/1

Y1 - 2018/1/1

N2 - This paper deals with the problem of measurement of the losses in Power GaN chips due to almost impossibility to measure the current under the penalty of adding parasitic inductances incompatible with the switching speeds of these components. For this purpose, this paper investigates the steady-state inverse heat conduction model (IHCM) developed from thermal simulations. The main objective is to derive the power dissipation in a GaN chip mounted on or embedded in a PCB from the surface temperature measured by infrared thermography or thermal sensors. In this problem, the power devices losses are the solutions of the inverse problem of the temperature-to-power mapping. IHCM here is based on simple observations from real PCB structures with GaN chip embedded in PCB substrates. The error of the resultant power map with thermal noises is minimized by Least-Square Optimization, which transforms the direct inverse problem into a constrained optimization problem.

AB - This paper deals with the problem of measurement of the losses in Power GaN chips due to almost impossibility to measure the current under the penalty of adding parasitic inductances incompatible with the switching speeds of these components. For this purpose, this paper investigates the steady-state inverse heat conduction model (IHCM) developed from thermal simulations. The main objective is to derive the power dissipation in a GaN chip mounted on or embedded in a PCB from the surface temperature measured by infrared thermography or thermal sensors. In this problem, the power devices losses are the solutions of the inverse problem of the temperature-to-power mapping. IHCM here is based on simple observations from real PCB structures with GaN chip embedded in PCB substrates. The error of the resultant power map with thermal noises is minimized by Least-Square Optimization, which transforms the direct inverse problem into a constrained optimization problem.

KW - Inverse model

KW - Least-square optimization

KW - PCB substrates

KW - Power losses estimation

KW - Temperature-to-power mapping

UR - http://www.scopus.com/inward/record.url?scp=85066976625&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85066976625&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85066976625

SN - 9783800746460

T3 - PCIM Europe Conference Proceedings

SP - 939

EP - 943

BT - PCIM Europe Conference Proceedings

A2 - Scharf, Achim

PB - Mesago PCIM GmbH

T2 - International Exhibition and Conference for Power Electronics, Intelligent Motion, Renewable Energy and Energy Management, PCIM Europe 2018

Y2 - 5 June 2018 through 7 June 2018

ER -