基于灰色关联和证据理论的故障诊断方法.docVIP

基于灰色关联和证据理论的故障诊断方法林云哈尔滨工程大学信息技术研究所哈尔滨 150001本文在证据理论的基础上, 结合信息熵和灰色关联算法, 提出了一种新的机械故障诊断方法。该方法从信息融合的思想出发, 首先依据反映机械故障的信息熵特征, 获得基于信息熵的故障诊断标准特征向量。接着采用灰色关联理论建立证据理论的基本概率赋值函数, 提出了利用证据理论对单传感器多测量周期证据时域融合和多传感器证据空域融合相结合的时空二级融合算法, 最后以基本可信数的决策方法作为故障模式识别依据。通过旋转机械故障的典型实例证明, 基于灰色关联和证据理论的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。灰色关联证据理论信息熵机械故障时空融合TN301.6　　　文献标识码: A　　　国家标准学科分类代码: 520.2099 Fault diagnosis method based on gray correlation and evidence theory Lin Yun Gao Lipeng (Institute of Information Technology, Harbin Engineering University, Harbin 150001, China) Abstract: Based on the Evidence Theory, combing with Information Entropy and Gray Correlation Algorithm, a new method of machinery fault diagnosis is presented in this paper. It starts from information fusion, firstly based on Information Entropy feature of machinery fault, the standard feature vectors for fault diagnosis is built. Secondly, the Basic Probability Assignment Function (BPAF) of Evidence Theory is built by Gray Correlation Theory, and then a space-time second-level fusion algorithm based on Evidence Theory is proposed, which includes the time domain fusion of single sensor evidence of mutli-measuring period and the space domain fusion of multisensor. Finally, a decision-making method based on the basic probability number is used as the fault model recognition. The typical instance of rotational machinery proves that the new machinery fault diagnosis method based on Gray Correlation and Evidence Theory is valid and feasible for recognizing fault models in quantification. Keywords: gray correlation; evidence theory; information entropy; machinery fault; space-time fusion 1 引 言 由于产生机械故障的因素众多, 一种故障可用不同的特征指标来描述, 同一种症状又常常是几种故障共同作用的结果。严格说来, 检测量与故障特 征之间, 故障特征与故障源之间都是一种非线性映射。因此, 故障的多样性、不确定性和各种故障之间联系的复杂性构成了故障诊断技术上的难点。仅靠单一传感器故障特征量和诊断方法无法完成诊断任务, 比较合理的方法就是采用信息融合技术, 进行多传感器多征兆的信息融合诊断。 本文从基于证据理论的信息融合思路出发, 利用反映机械故障的信息熵特征, 建立基于信息熵的故障诊断标准特征向量。利用这些标准特征向量作为证据体, 以灰色关联算法来建立基本概率赋值函数。在此基础上, 利用一种改进的时空域证据融合模型, 将这些证据进行更加有效地组合。最后, 采用基于基本可信数的决策来进行故障诊断