噪声与振动控制
譟聲與振動控製
조성여진동공제
NOISE AND VIBRATION CONTROL
2015年
4期
6-10
,共5页
振动与波%灰色关联度分析方法%时频域特征值%灰色神经网络模型%关联度计算%故障诊断
振動與波%灰色關聯度分析方法%時頻域特徵值%灰色神經網絡模型%關聯度計算%故障診斷
진동여파%회색관련도분석방법%시빈역특정치%회색신경망락모형%관련도계산%고장진단
vibration and wave%grey correlation analysis method%characteristic value in time and frequency domain%gray neural network%correlation algorithm%fault diagnosis
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。
在分析常規關聯度計算方法存在問題的基礎上，提齣一種改進的灰色關聯度對柴油機故障診斷的方法。針對柴油機故障的特徵，對經過降譟後的振動信號，提取時域頻域特徵值相結閤作為特徵嚮量，分彆通過改進的灰色關聯度方法、常規灰色關聯度方法以及灰色神經網絡模型對待檢測特徵嚮量和標準模式嚮量進行關聯度計算。對結果進行分析，得知改進的灰色關聯度分析方法剋服常規灰色關聯度容易誤判的缺陷，同時驗證改進的灰色關聯度方法大大的提高柴油機故障診斷的精度，說明該方法是一種有效可行的方法。
재분석상규관련도계산방법존재문제적기출상，제출일충개진적회색관련도대시유궤고장진단적방법。침대시유궤고장적특정，대경과강조후적진동신호，제취시역빈역특정치상결합작위특정향량，분별통과개진적회색관련도방법、상규회색관련도방법이급회색신경망락모형대대검측특정향량화표준모식향량진행관련도계산。대결과진행분석，득지개진적회색관련도분석방법극복상규회색관련도용역오판적결함，동시험증개진적회색관련도방법대대적제고시유궤고장진단적정도，설명해방법시일충유효가행적방법。
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.
