柔性机械臂振动控制中的压电传感器故障诊断  

Faults diagnosis of piezoelectric sensor for flexible arm vibration control

作　　者：马天兵[1] 杜菲[1] 熊能[1] 钱星光[1] 

机构地区：[1]安徽理工大学机械工程学院机电系,淮南232001

出　　处：《电子测量与仪器学报》2014年第12期1408-1413,共6页Journal of Electronic Measurement and Instrumentation

基　　金：国家自然科学基金资助项目(51305003);安徽省高校优秀青年人才基金重点资助项目(2012SQRL045ZD)

摘　　要：压电传感器常被应用于结构的振动控制中,为了解决压电传感器故障给控制系统的带来的不稳定问题,根据压电柔性臂振动用压电传感片故障的特点,提出运用小波包变换和RBF神经网络的故障诊断方法。首先运用小波包分解和重构原理将传感器输出信号分解到不同频段中,提取每个频段的能量作为状态监测的特征向量,作为RBF网络的输入,然后利用最佳的RBF神经网络进行压电传感器故障分类。实验结果表明该方法具有良好的非以此线性跟踪能力,置信度达到90%,为后续振动容错控制研究奠定良好的基础。In order to solve the problem of controller instability for piezoelectric sensor faults in the vibration control of structure,according to the faults character of piezoelectric sensor for vibration control of piezoelectric flexible arm,a new diagnosis method based on wavelet packet transform and RBF neural network is proposed to detect and identify the sensor fault. By using wavelet packet decomposition and reconstruction,the sensor fault signals are decomposed in different frequency bands,and the energy of every band is used as the eigenvector of condition monitoring as well as input of RBF neural network. The classification of sensor fault is conducted by using the best RBF neural network. The experiment results prove that the method has good tracking ability of nonlinear system and its confidence can reach 90%,so that it can lay good foundation of follow-up fault-tolerant vibration control study.

关 键 词：压电传感器 小波包变换 神经网络 故障诊断 

分 类 号：TP206[自动化与计算机技术—检测技术与自动化装置]