摘要
特征提取是机械设备故障诊断成功的关键因素,将直接决定故障诊断的实施。针对传统旋转机械单通道故障诊断的不足,结合故障诊断实时性差的问题,提出了一种基于全矢谱理论和蚁群算法的机械故障特征提取方法。运用全矢谱技术提取出原始振动信号的特征向量,再运用改进的蚁群算法对特征参数进行约简和选择,使所建立的故障模式由少数几个特征给予有效的表达,以提高故障诊断的准确性和快速性。实验证明该方法是有效的。
Extractioh of eigenvector is key taehe on successful fault diagnosis ,and it affects the execution of fault diagnos is directly. Due to the insufficiency of traditional rotary machinery fault diagnosis with single channel signal and the deficiency of rapidity. A new rotary machinery fault feature extraction method combining full vector spectrum theory with ant colony algorithm is proposed. Extracting the eigenvector of original vibration signal with the technology of full vector spectrum, and then reducing the feature with ant colony algorithm,so the fauh mode is effectively expressed by several feature,thus the accuracy and the rapidity of fault diagnosis was enhanced. The experiment result shows that this method is effective.
出处
《机械设计与制造》
北大核心
2009年第10期35-37,共3页
Machinery Design & Manufacture
基金
国家自然科学基金资助项目(50675209)
河南省杰出人才创新基金(0621000500)
关键词
特征提取
全矢谱
蚁群算法
故障诊断
Feature extraction
Full vector spectrum
Ant colony algorithmt
Fault diagnosis
