基于模糊C均值算法的齿轮泵故障诊断技术

Diagnosis Technics of Gear Pump Fault Based on the Fuzzy C-means Clustering Algorithm

以CB-KP63齿轮泵为研究对象,提出了利用小波包频带能量提取齿轮泵的信号特征,用模糊C均值聚类算法得到齿轮泵的故障模式,然后用模糊贴近度进行故障模式的识别。实验结果表明,在此算法下,齿轮泵4种不同工作状态下的振动信号具有明显的可分性,并且该诊断方法对不同类型的齿轮泵故障诊断、维修保养等都具有一定的实用性。

Wavelet packet frequency band energy was used to extract characteristic of the signal for CB-KP63 gear pump. Fuzzy C-means clustering algorithm was used to obtain the failure mode of gear pump. Fuzzy nearness degree was proposed to identify different failure mode. Experimental results show that based on this algorithm, the vibration signals of gear pumps can be clearly distinguished from four different conditions. This method has a certain practicality for different types of gear pump in fault diagnosis and maintenance.