Simplified Estimation of Junction Temperature Fluctuation at the Fundamental Frequency for IGBT Modules Considering Mission Profile

Xiping Wang,Shengxue Tang,Zhigang Li,Fang Yao

doi:10.1109/access.2019.2947346

Abstract

The junction temperature at the fundamental frequency cannot be ignored in a lifetime evaluation of insulated-gate bipolar transistors (IGBTs) with a long-term mission profile. Therefore, it is very important in terms of calculation speed and accuracy to simplify the loss curve when calculating the thermal fluctuation at the fundamental frequency. This paper proposes a mathematical analysis method for the junction temperature fluctuation at the fundamental frequency based on an equivalent sinusoidal half-wave loss. The dynamic and static parameters of the devices are tested by an experimental platform, and accurate device loss models are established through a case study of a 1.5 MW direct-drive wind turbine grid-connected model. The accuracy of the proposed calculation model is compared with that of a time-domain simulation and the two-step loss pulse method. Considering different output frequencies, the accuracy of the proposed method is further discussed. Based on actual wind speed data, the proposed method is used to calculate the junction temperature of an IGBT module in a grid-side converter. The results show that the proposed method can improve the accuracy of the reliability evaluation of wind power converters.

Highlights

In recent years, wind energy has developed rapidly as a clean renewable energy source, and the capacity of wind turbines has increased continuously
For actual wind speed data, a quantitative comparison is made with the twostep loss pulse method, which further validates the effectiveness of the proposed method
The validity of the calculation of the junction temperature fluctuation is verified for output power frequencies through the grid-connected inverter model

Summary

INTRODUCTION

Wind energy has developed rapidly as a clean renewable energy source, and the capacity of wind turbines has increased continuously. Reference [22] considered the inherent DC bias characteristics of a modular multilevel converter (MMC) and proposed a mathematical iterative method based on a virtual sinusoidal half-wave loss approximation to calculate the junction temperature fluctuation of power devices in the MMC, and the validity of the method was verified. By inputting the equivalent rectangular pulse loss of the sinusoidal half-wave loss into the calculation model of the steady-state junction temperature fluctuation, the maximum fluctuation value Tjmax can be obtained This method is applicable to the junction temperature calculation for the grid side and the machine side in the converter. Taking the grid-side inverter as an example, the junction temperature of the chip is evaluated and iteratively calculated when calculating the average conduction losses of the IGBT and the diode to better consider the electrothermal interaction characteristics under operating conditions. The average junction temperature and average loss of the power devices are obtained under the corresponding working conditions

CALCULATION MODEL OF THE THERMAL FLUCTUATION AT THE FUNDAMENTAL FREQUENCY

DYNAMIC AND STATIC PARAMETER TESTING

FAST ESTIMATION OF THE JUNCTION TEMPERATURE WITH ACTUAL WIND SPEED DATA

Findings

CONCLUSION