摘要
针对小口径火炮自动机工作时产生的短时冲击信号,提出一种将局域波分解与信息熵相结合提取特征量,并利用Elman神经网络进行故障识别的诊断方法。首先运用具有自适应特性的局域波对振动信号进行分解得到IMF分量,再接着利用信息熵理论提取IMF信息熵、局域波能谱熵及能矩谱熵作为故障特征量,最后将特征向量输入Elman神经网络进行故障分类识别。实验结果表明:该方法能准确,有效地识别故障。
As short-term impact signal will produce when small caliber gun＇s automaton works, a diagnostic method for this case has been proposed with the local wave decomposition and information entropy combined extract feature vectors and using Elman neural network recognition fault. Firstly, the local wave with self adaptive feature is used to break down the vibration signal into a series of IMF components. Then it takes advantage of the information entropy theory to extract the IMF information entropy, local wave energy spectrum entropy and energy moment spectrum entropy as the fault feature. Finally, the feature vector is input to the Elman neural network for fault classification and recognition. The experimental results show that the method can identify the fault accurately and effectively.
出处
《中国测试》
CAS
北大核心
2014年第1期115-118,136,共5页
China Measurement & Test
基金
国家自然科学基金项目(51175480)
关键词
人工智能
高速自动机
故障诊断
信息熵
局域波
artificial intelligence
high-speed automaton
fault diagnosis
information entropy
local wave
