证据熵在旋转机械故障诊断中的应用

Fault Diagnosis of Rotating Machinery Based on Evidence Theory of Evidence Entropy

针对旋转机械故障诊断中利用证据理论进行多征兆域融合时各证据的重要性不同,提出利用证据熵来评价各证据的重要性,即各证据在融合过程中的权重。对得出的证据进行加权调整,用D-S组合规则对加权调整后的证据进行融合获得最终的诊断结果。实例分析结果表明,这种方法可以用来评价证据的重要性,且由于其在进行融合时减少了证据间的冲突,所以该方法能够在旋转机械故障诊断应用中取得较好的效果。

Considering the importance difference of evidence when applying the evidence theory to fuse multi-symptom-domain during fault diagnosis of rotating machinery,this paper adopted the evidence entropy to estimate the importance of evidence,i.e.the weight of each piece of evidence.Then,the evidence was adjusted according to the different weights,and the weighted adjusting evidence was fused by the D-S combination rule,so that the final diagnosis result was obtained.Finally,by a practical example,the result shows that the method can estimate importance of evidence and reduces the contradictions among different pieces of evidence.So,this method is effective in diagnosing faults of rotating machinery.