摘要
提出了基于小波矩特征和模糊核聚类算法的示功图故障诊断方法。通过边缘检测和形态学细化的方法完成示功图的图像分割,采用极坐标下小波不变矩算法提取示功图的形状特征,通过参数选择确定12个小波矩特征量,将特征量输入到模糊核聚类分类器中进行故障类型的分类识别,得到了良好的实验效果,验证了该算法对于示功图故障诊断的有效性。
Based on wavelet moment and kernel fuzzy C-means,it proposes the fault diagnosis method of indicator diagram.Through edge detecting and morphological refining to complete the image segmentation of indicator diagram,it extracts the shape feature of the indicator diagram with wavelet moment algorithm under polar coordinates,determines 12 wavelet moment characteristics from parameter selection.Inputting above algorithm to the fuzzy kernel clustering classifier,it identifies the fault type classification,obtains the good experimental result.The result shows that this method can recognize the fault types of the indicator diagram effectively.
出处
《机械设计与制造工程》
2016年第7期80-83,共4页
Machine Design and Manufacturing Engineering
关键词
示功图
故障诊断
图像分割
小波矩
模糊核聚类
indicator diagram
fault diagnosis
image segmentation
wavelet moment
kernel fuzzy C-means
