摘要
绝缘栅双极型晶体管(IGBT)开关过程的di/dt和du/dt是影响换流器电磁干扰(EMI)水平的主要因素。IGBT的寄生振荡是高频EMI的重要组成部分,振荡频点处会出现EMI峰值。该文提出一种考虑寄生振荡的IGBT分段暂态模型,分析回路寄生参数和器件非线性电容对开关特性的影响,分别计算不同阶段的电流和电压变化率。搭建二极管钳位感性负载测试平台,获取IGBT的电流和电压波形,分析对比分段模型和实际波形的频谱特性。最后,通过实验验证了模型中的振荡过程是影响电流频谱特性的关键,且采用器件电容C_(gc)的三段等效模型可以显著提高电压频谱预测的准确度。该文提出的模型提高了IGBT干扰源频谱的预测准确度,可用于评估实际换流器发射的EMI水平。
The di/dt and du/dt of the insulated gate bipolar transistor(IGBT)switching process are the major factors affecting the electromagnetic interference(EMI)level of the converter.The parasitic oscillation of IGBT is an important part of high frequency EMI,and the EMI peak appears at the oscillation frequency.In this paper,an IGBT segmented transient model considering parasitic oscillations was presented.The effects of parasitic parameters and device nonlinear capacitance on switching characteristics were analyzed.The voltage and current change rates at different stages were calculated respectively.Then,a diode-clamped inductive load test platform was built.The current and voltage waveforms of IGBT were obtained,and the spectrum characteristics of the segmented model and the actual waveform were analyzed and compared.Finally,it is verified through experiments that the oscillation process in the model is the key factor affecting the current spectrum characteristics,and the three-stage equivalent model of Cgc can significantly improve the accuracy of voltage spectrum prediction.The proposed model improves the prediction accuracy of the IGBT interference source spectrum and can be used to evaluate the EMI level of the actual converter.
作者
黄华震
仝涵
王宁燕
卢铁兵
Huang Huazhen;Tong Han;Wang Ningyan;Lu Tiebing(State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources North China Electric Power University,Beijing 102206,China;Fujian Electrical Power Research Institute,Fuzhou 350003,China)
出处
《电工技术学报》
EI
CSCD
北大核心
2021年第12期2434-2445,共12页
Transactions of China Electrotechnical Society
基金
国家电网公司科技资助项目(52130419000M)。
关键词
绝缘栅双极型晶体管
分段暂态模型
寄生振荡
频谱
Insulated gate bipolar transistor(IGBT)
segmented transient model
parasitic oscillations
spectrum
