| IGBT机理建模及其基于神经网络的参数辨识方法 |
| |
| 引用本文： | 孙跃,谭晶晶,唐春森.IGBT机理建模及其基于神经网络的参数辨识方法[J].西南交通大学学报,2015,28(6):1143-1149,1163. |
| |
| 作者姓名： | 孙跃 谭晶晶 唐春森 |
| |
| 基金项目： | 国家863计划资助项目(2015AA010402)中央高校基本科研业务费专项基金资助项目(106112015CDJXY170002)重庆市基础与前沿研究计划一般项目(cstc2013jcyjA0235)重庆市研究生科研创新项目(CYS14026) |
| |
| 摘 要： | 为获取专用的绝缘栅双极型晶体管(insulated gate bipolar transistor,IGBT)模型,实现IGBT电路的针对性优化,在Hefner模型的基础上,对IGBT的开通和关断状态进行了机理建模,并重点分析了其暂态过程.在此基础上,提出了基于神经网络优化算法对模型参数进行辨识的方法,获得了单个IGBT元件的机理模型.以一个FGA25N120型的IGBT为例,进行了仿真和实验研究,通过仿真与实验结果的对比,拟合优度达0.9,验证了本文所提机理模型的正确性及基于神经网络辨识所得参数的精确性.
|
| 关 键 词： | 绝缘栅双极型晶体管 神经网络 机理模型 参数辨识 开关暂态 |
| 收稿时间： | 2015-05-23 |
Physical Modeling of IGBT and Its Parameter Identification Method Based on Neural Network |
| |
| SUN Yue,TAN Jingjing,TANG Chunsen.Physical Modeling of IGBT and Its Parameter Identification Method Based on Neural Network[J].Journal of Southwest Jiaotong University,2015,28(6):1143-1149,1163. |
| |
| Authors: | SUN Yue TAN Jingjing TANG Chunsen |
| |
| Abstract: | In order to obtain the special model of insulated gate bipolar transistor (IGBT), and realize targeted optimization of the IGBT circuit, the IGBT during its on and off state was modeled separately based on Hefner model and the transient was analyzed. Meanwhile, the model parameter identification method based on neural network optimization algorithm was proposed to obtain the physical model of a single IGBT component. Finally, the parameters of an IGBT of FGA25N120 were used as an example. By the comparison of simulation and experimental results, the goodness of fit of the model was 0.9, which verifies the correctness of the proposed physical model and the accuracy of parameters identified by neural network. |
| |
| Keywords: | |
|
| 点击此处可从《西南交通大学学报》浏览原始摘要信息 |
|
点击此处可从《西南交通大学学报》下载全文 |
|
