基于信息熵贴近度的旋转机械故障诊断

基于信息融合的思想，研究了反映振动能量的旋转机械故障状态的各种信息熵特征，如奇异谱熵、功率谱熵、小波空间状态特征谱熵和小波能谱熵．通过转子试验，给出了旋转机械的不平衡、不对中、支座松动、轴裂纹典型故障下的各信息熵的变化范围．根据越相似的模式间距离越短原理，提出采用贴近度来进行模式识别的方法．首先利用贴近度原理和熵带构建了信息熵贴近度模型，其次计算出待识别状态与各典型故障之间的信息熵贴近度值，则对应于待识别状态之间的信息熵贴近度最大的即为待识别状态的故障模式，最后通过实例描述了基于信息熵贴近度的旋转机械故障诊断方法的可行性．

Rotating machinery fault diagnosis based on close degree to information entropy

From information fusion, the various information entropy features, which explicate the faults from vibration energy in rotating machineries, such as singular spectrum entropy, power spectrum one, wavelet space state feature and wavelet power spectrum ones, were identified. By the experiment on the rotor, the range of the value of the various entropies are set up, which describes the typical faults of the rotating machinery such as imbalance, misalignment, pedestal looseness, shaftr crack. According to the thought that the more short of the distance, the more similar among the models, it is presented that the pattern is recognized by the close degree. Firstly, the model of the close degree to information entropy is constructed by the theory of close degree and the entropy rang. Secondly, the value of the close degree to information entropy between the unknown state and the different typical fault state, so, the fault state with the maximal value of the close degree to information entropy between the unknown state is the state of the unknown state. At last, the instance described the feasibility of the method of rotating machinery fault diagnosis based on the close degree to information entropy.