论文部分内容阅读
从信息融合的思路出发,建立反映振动能量的旋转机械故障状态的信息熵特征,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。通过试验,建立了旋转机械典型故障下的信息熵期望值,即获得基于信息熵的故障诊断标准特征向量。由于传感器的不确定性和故障的多样性,提出采用D-S证据理论来对4种信息熵进行信息融合。根据越相似模式间的距离越短的思路,提出采用信息熵贴近度来建立证据理论的基本可信度分配,以基于基本可信数的决策方法来作为故障模式识别方法。通过实例计算,证明基于信息熵贴近度和证据理论的旋转机械故障诊断方法是故障模式定量识别的一种可行的新方法。
Based on the idea of information fusion, the information entropy features of rotating machinery such as singular spectrum entropy, power spectral entropy, state-of-feature spectrum entropy and wavelet energy spectrum entropy of rotating machinery are established. Through experiments, the expectation of information entropy under typical faults of rotating machinery is established, that is, the fault diagnostic standard eigenvector based on information entropy is obtained. Due to the uncertainty of sensors and the diversity of faults, this paper proposes to use D-S evidence theory to fuse four kinds of information entropy. According to the idea that the distance between similar patterns is shorter, the paper puts forward a method to establish the basic credibility distribution based on the closeness of information entropy and the decision method based on basic credible. Through case calculation, it is proved that the method of rotating machinery fault diagnosis based on information entropy closeness and evidence theory is a feasible new method for quantitative identification of fault modes.