摘要
对于目前传感器设备故障诊断方法在诊断过程中会导致实际生产中弱故障诊断不灵敏,无法有效识别故障,并且充分考虑实际生产过程中的特点,提出基于FPGA的传感器故障诊断算法设计。在不同残差空间中分析故障,使系统对于弱故障诊断的精准性得到提高。使用迭代算法对主元分析算法模型进行更新,利用残差空间平方加权预测误差变量的重构对故障进行确定,从而对传感器故障进行在线诊断。最后,通过企业实际数据实现此方法的在线故障诊断实验。实验结果表明,此种方法能够有效诊断传感器设备故障。
For the current sensor equipment fault diagnosis methods in the process of diagnosis may lead to insensitivity to weak fault diagnosis in the actual production,and ineffectiveness to the fault identification,the design of sensor fault diagnosis algorithm based on FPGA is proposed by fully considering the characteristics of the actual production process. The fault analysis in different residual spaces improves the accuracy of the system for weak fault diagnosis. The principal component analysis (PCA)model is updated by iterative algorithm,and the reconstruction of error variable is predicted by the residual space square weighting to determine the fault,so that the sensor fault can be diagnosed online. The online fault diagnosis experiment was carried out for this method according to the actual data of an enterprise. The experimental results show that this method can effectively diagnose sensor equipment faults.
作者
李小明
李萌欣
LI Xiaoming;LI Mengxin(Henan Radio & Television University,Zhengzhou 450046,China)
出处
《现代电子技术》
北大核心
2019年第19期178-181,186,共5页
Modern Electronics Technique
关键词
传感器故障诊断
故障识别
弱故障诊断
故障在线诊断
故障分析
模型更新
sensor fault diagnosis
fault recognition
weak fault diagnosis
online fault diagnosis
fault analysis
model update
