基于小波包-神经网络方法的支撑座连接螺栓松动损伤诊断的实验研究  

Experimental Study on Diagnosing the Attachment Bolt Looseness in a Clamping Support Based on Wavelet Packet Transformation and Neural Network

作　　者：董广明[1] 陈进[1] 雷宣扬[1] 宁佐贵[2] 王东升[2] 王雄祥[2] 

机构地区：[1]上海交通大学振动冲击噪声国家重点实验室,上海200240 [2]中国工程物理研究院结构力学研究所,绵阳621900

出　　处：《机械科学与技术》2006年第1期102-106,共5页Mechanical Science and Technology for Aerospace Engineering

基　　金：国家自然科学基金项目(10176014)资助

摘　　要：在某导弹支撑座模型宽带随机振动实验的基础上,针对其连接螺栓松动所产生的支撑座结构响应的非平稳特性,采用小波包分析的方法得到缩减的信号特征;然后利用BP神经网络的模式分类功能,进行了螺栓松动程度的损伤识别研究。实验结果表明,小波包结合神经网络的方法可以有效地识别该支撑座连接螺栓的松动程度。On the basis of broad-band random vibration test on a clamping support in a missile, wavelet packet analysis method was used to obtain the reduced features of structural response signals, which have non-stationary characteristics due to the attachment bolt looseness. Then damage identification of severity of bolt looseness was studied by utilizing the pattern classification function of the BP neural network. Test results show that the combination of wavelet packets and neural networks can effectively identify the severity of bolt looseness in the clamping support.

关 键 词：宽带随机振动实验 支撑座结构 螺栓松动 小波包 BP神经网络 

分 类 号：TH113.1[机械工程—机械设计及理论]