基于IGBT特征分析的地铁车辆逆变器故障研究
吴浩王泉王睿轶
Study on Metro Vehicle Inverter Fault Based on IGBT Characteristic Analysis
WU HaoWANG QuanWANG Ruiyi
-
作者信息:上海地铁电子科技有限公司，200233，上海
-
Affiliation:Shanghai Metro IT Co., Ltd., 200233, Shanghai, China
-
关键词:
-
Key words:
-
DOI:10.16037/j.1007-869x.2020.09.012
-
中图分类号/CLCN:U264.3+7
-
栏目/Col:研究报告
摘要:
针对地铁车辆功率逆变器，提出一种改进的基于SVM（支持向量机）的故障状态识别方法。从研究IGBT（绝缘栅双极型晶体管）结温与故障的关系入手，采用多分辨率小波包分解对主电路IGBT特征状态进行提取，特征降维后获得SVM训练集，经SVM学习训练后得到状态分类模型。仿真结果表明:分类模型较为接近实际情况。最后通过试验验证了该方法能有效地识别单个及多个IGBT软故障和硬故障，实现功率逆变器状态识别。
Abstracts:
In view of metro vehicle power inverter, a fault status recognition method based on the improved SVM (support vector machine) is proposed. Proceeding from the relationship between junction temperature and fault of IGBT (insulated gate bipolar transistor), a multiple resolution wavelet packet decomposition method is used to extract the feature status in main circuit of IGBT. After feature dimension reduction, the SVM training set is formed, and the status classification model is obtained after SVM learning and training. The simulation results show that the accuracy of classification model is close to the reality. Finally, through real tests, this method is verified to be effective in identifying single and multiple IGBT soft and hard faults, thus achieving power inverter status recognition.
- 上一篇: 高铁物流运输模式及其可行性
- 下一篇: 城市数字化转型给轨道交通带来的机遇和挑战
过刊浏览
资料下载
友情链接