论文部分内容阅读
在IGBT寿命预测研究中,应用雨流法进行器件所承受应力周次统计可取得较好的效果;但其必须基于完整的应力谱,故存在数据量大、无法满足实时应用需求等弊端。文章提出一种改进的雨流算法,其针对应力谱测试数据流,先经数据清洗处理转换成极值数据,再进行应力循环辨识和计数,最终得到实时的动态应力幅谱和均值结果输出。仿真和工程化应用结果表明,该方法的研究可为实现IGBT健康状态及应用寿命的实时在线测算提供技术支撑,具备一定的应用价值。
In the research of IGBT lifetime prediction, the rainstorm method can get better effect on the weekly statistics of the stress of the device, but it must be based on the complete stress spectrum, so there is a large amount of data, which can not meet the real-time application needs and other shortcomings. In this paper, an improved rain flow algorithm is proposed, which is used to test the data flow of stress spectrum. After the data is cleaned, the data is transformed into the extreme value data, and then the stress cycle is identified and counted. Finally, the real-time dynamic stress spectrum and the average output are obtained. Simulation and engineering application results show that this method can provide technical support for real-time on-line measurement of IGBT health status and service life, and has certain application value.