摘要
针对目前设备状态报警线设置所存在的缺点,即只依据单通道信息及设定的报警阈值和设备实际运行状态变化无关,本文提出了一种基于多通道全信息技术的自适应报警方法。该方法利用概率神经网络构建设备运行状态模型,根据多通道历史数据确定报警值并设置报警线。实验表明,该方法是可行的,而且是有效的,用该方法设置的设备状态报警线能够随设备运行而做自适应调整,对指导工业现场的设备监测具有现实意义。
There exist the shortcanings that warning threshold values are determined only in accordance with single-channel information and have nothing to do with the changes in the working conditions of machines. For this rearon, the paper proposed an adaptive warning method based on multi-channel full information technique. With this method, the machines＇ working condition model is built through using the probabilistic neural network, and their warning values are determined and warning lines are set according to multi-channel historical data. An experiment indicates that the method is feasible and effective and that the warning lines thus set can adjust themselves in adaptation to the machines＇ working conditions, shedding light on the machine monitoring at industrial sites.
出处
《机械科学与技术》
CSCD
北大核心
2006年第12期1499-1502,共4页
Mechanical Science and Technology for Aerospace Engineering
基金
河南省杰出人才创新基金项目(0621000500)资助
关键词
全信息技术
自适应报警方法
概率神经网络
full information technique
adaptive alarm method
probabilistic neural network
