基于傅里叶描述子的示功图诊断方法研究  

Fault Diagnosis Method for Indicator Diagram Based on Fourier Descriptors

作　　者：丛蕊[1] 乔磊[1] 张威[1] 

机构地区：[1]东北石油大学机械科学与工程学院

出　　处：《化工机械》2013年第3期285-288,308,共5页Chemical Engineering & Machinery

基　　金：国家"十二五"科技支撑计划项目(2011BAA02B01)

摘　　要：提出了基于傅里叶描述子提取轮廓线特征的示功图故障诊断方法。利用八邻域搜索和形态学方法提取示功图轮廓线,在此基础上计算轮廓线边界点序列的归一化傅里叶描述子,保证了图像对于旋转、缩放和平移的不变性,并将其作为特征量输入到距离分类器中,从而实现故障智能诊断。实验结果表明,该方法可以有效地识别出示功图的不同故障类型。The fault diagnosis method of extracting the contour line features based on Fourier descriptors was proposed. Through using eight-neighborhood search and morphological refining to extract contour line, the Fourier descriptor of contour line boundary point sequence was calculated to ensure image＇ s constantness in rotation, zooming and panning; and inputting the Fourier descriptor as the feature to distance classifier can realize intelligent diagnosis of the fault. The experimental result shows that this method can recognize the fault types of the indicator diagram effectively.

关 键 词：往复机械 故障诊断 形态学细化 八邻域搜索 傅里叶描述子 

分 类 号：TQ050.7[化学工程]