摘要
提出了一种汽轮机故障监测及诊断系统的数据预处理的方法。该方法在采集部分使用了一种先进行高密度采集而后进行插值处理的数据采集方法,这种数据采集方法可以充分地利用现有硬件的强大性能,在转速波动比较大时也能采集到比较理想的数据。同时,在此方法中设置有故障的预警功能。此功能利用了基于统计学原理的状态评判方法来实现,与常用的方法(如能量法)相比,具有反应速度较快、灵敏度较高的优点。
A method of data preprocessing method about the condition monitoring and fault diagnosis system for turbine is proposed. This method uses an advanced high density sampling and interpolation algorithm, which can effectively gather the samples even under the large fluctuation of the rotate speed. In addition, alert effect is mounted on this system. This alarm employs a state decision based on statistics theories. Compared with the traditional method such as energy method, this method has the rapid response speed and high sensitivity.
出处
《汽轮机技术》
北大核心
2005年第2期134-136,113,共4页
Turbine Technology
关键词
汽轮机
状态监测
故障诊断
预处理
steam turbine
condition monitoring
fault diagnosis
preprocessing
