摘要
为了解决煤矿压风机监控系统关键传感器的故障检测与分离问题,提出了一种基于主元分析模型的传感器故障诊断方法。该方法可辨识系统中相关性较高的若干传感器,并为之建立主元分析模型。根据所建立的模型,利用平方预报误差(SPE)判断系统中是否有传感器发生故障;利用SPE贡献图定位故障传感器。系统经阶跃偏差故障试验和漂移故障试验,结果表明,将主元分析法应用于煤矿压风机监控系统传感器的故障检测与分离,可为压风机系统的正常运行提供有力保障。
In order to solve the key sensor fault detection and separation problems of the mine pressurized fan monitoring and control system,a sensor fault diagnosis method based on the main element analysis model was provided.The main element analysis model was established with the several sensors highly related in the identification system of the diagnosis method.According to the established model,the square prediction error was applied to judge any sensors with fault in the system.The contribution plot of the square prediction error was applied to position the fault sensor.The system had the step deviation fault test and the drift fault test and the results showed that the main element analysis method could be applied to the sensor fault detection and separation of the monitoring and control system for the mine pressurized fan and could provide the powerful protection to the normal operation of the mine pressurized fan system.
出处
《煤炭科学技术》
CAS
北大核心
2011年第6期82-85,共4页
Coal Science and Technology
基金
河南理工大学博士基金资助项目(B2010-48)
关键词
压风机监控系统
主元分析法
传感器
故障检测与分离
monitoring and control system of pressurized fan
main element analysis method
sensor
fault detection and separation
