旋转机械参数图形软形态学自适应边缘检测

Fuzzy soft morphology self-adaptive edge detection of parameter image for rotating machinery

针对旋转机械振动参数图形边缘特征提取困难问题,根据模糊软形态学理论,提出滤波增强处理方法及自适应边缘检测算子.在600 MW模化汽轮机转子试验台上进行转子正常运转、转子不平衡故障、转子不对中故障、汽流激振故障、轴承松动故障的实验研究.将得到的振动参数三维图形转化为二维灰度图形,对二维灰度图形进行模糊软形态学滤波增强处理和自适应边缘检测.结果表明,该方法在滤除参数图形中噪声的同时,可以有效地提取图形边缘特征,为旋转机械故障诊断提供了一种新的图形特征提取方法.

Aiming at the problem that the edge features of vibration parameter images for rotating machinery are difficult to be extracted, the filtering enhancement processing method and the self-adaptive edge detection operator are established according to fuzzy soft morphology theory. The 3d vibration parameter images of rotor＇s normal state, fault of unbalance, misalignment, steam exciting vibration and bearing pedestal looseness were obtained from the experiments on the modeling of 600 MW turbine rotor experimental bench. These 3d images were transformed to 2d gray-scale images, and these 2d gray-scale images are processed with fuzzy soft morphology filtering enhancement processing and self-adaptive edge detection. The results show that with this method the noise of parameter images can be filtered out and the edge features of images can be extracted effectively, so that a new method to extract image features for rotating machinery fault diagnosis is provided.