改进的灰色关联法在柴油机故障诊断中的应用

Application of Improved Grey Correlation Method to the Fault Diagnosis of Diesel Engines

在分析常规关联度计算方法存在问题的基础上,提出一种改进的灰色关联度对柴油机故障诊断的方法。针对柴油机故障的特征,对经过降噪后的振动信号,提取时域频域特征值相结合作为特征向量,分别通过改进的灰色关联度方法、常规灰色关联度方法以及灰色神经网络模型对待检测特征向量和标准模式向量进行关联度计算。对结果进行分析,得知改进的灰色关联度分析方法克服常规灰色关联度容易误判的缺陷,同时验证改进的灰色关联度方法大大的提高柴油机故障诊断的精度,说明该方法是一种有效可行的方法。

Based on the analysis of the problems in the conventional correlation algorithm, an improved grey correlation method was proposed for fault diagnosis of diesel engines. In this method, according to the characteristics of the diesel engine＇ s faults, the characteristic values in the time domain and frequency domain were extracted from the denoised vibration signals, and then combined as the characteristic vector. Using the improved grey correlation method, conventional grey correlation method and grey neural network model respectively, the relativity between the detection feature vectors and standard mode vectors was calculated. The result was analyzed and compared. It is shown that the improved grey correlation analysis method can overcome the defects of misjudgment in conventional grey correlation algorithm, and greatly raise the precision of the diesel engine fault diagnosis. So, this method is an effective and feasible method.