计算机控制系统中传感器故障诊断方法仿真  

Sensor Fault Diagnosis Method Simulation in Computer Control System

作　　者：杨波[1] 

机构地区：[1]齐齐哈尔大学计算中心,黑龙江齐齐哈尔161006

出　　处：《计算机仿真》2014年第3期389-391,430,共4页Computer Simulation

基　　金：齐齐哈尔大学青年教师科研启动支持计划项目(2012K-Z05)

摘　　要：研究计算机控制系统传感器故障准确诊断问题。计算机控制系统中的电子器件多工作在高压环境下,传感器工作强度很大,但其本身的高精密性决定了其故障特征极信号容易受到干扰。传统的根据故障频率波进行传感器诊断的方法只能依靠输出波形判断缓变故障,而干扰的故障波形在频率中无法形成较为特殊的突变信号,造成传感器故障很难识别。提出了一种故障信息簇提升聚类的计算机控制系统的传感器故障检测方法。利用提升小波变换方法,对故障状态信号进行滤波处理,去除信号中的干扰噪声。采用故障信息簇聚类方法,对控制系统状态信号进行聚类处理,在可疑区域内获取传感器故障信号。实验结果表明,改进后的算法进行计算机控制系统智能故障检测,能够有效提高故障检测的准确性。The accuracy of computer control system fault diagnosis method was researched. Electronic devices in computer control system often work in a high-pressure environment, working strength of sensor is strong, and some signs of the fault characteristic are not obvious. Traditional method used to diagnose sensor according to the fault fre- quency wave, can only rely on the output waveform to judge fault, disturbed fault waveform can not form relatively specific mutation signal in frequency. In order to avoid the above defects, an intelligent fault detection method of computer control system was put forward based on fault information clustering algorithm. A lifting wavelet transform method was applied to proceed filtering processing for fault state signal and remove the interference noise in the sig- nal. The information clustering method of state signal of control system in clustering process was used to obtain the fault signals. The experimental results show that the algorithm presented in this paper for intelligent fault detection in computer control system, can effectively improve the accuracy of fault detection.

关 键 词：计算机控制 传感器 故障检测 信息簇聚类 

分 类 号：TP187.6[自动化与计算机技术—控制理论与控制工程]