학술논문
基于振动信号PSD-SVM方法的不定负荷下柴油机气阀间隙异常故障诊断 / Fault diagnosis of the gas valve of a diesel engine under uncertain load based on PSD-SVM
Document Type
Academic Journal
Author
Source
振动与冲击 / Journal of Vibration and Shock. 43(2):299-305
Subject
Language
Chinese
ISSN
1000-3835
Abstract
针对许多基于振动信号的故障诊断方法在不同负荷下的诊断不全面的问题.提出了一种基于功率谱密度(power spectral density,PSD)与支持向量机(support vector machine,SVM)的故障诊断方法.该方法将振动信号功率经过滑动平均滤波(moving average filter,MAF)处理,计算样本中每个周期的标准化信号的功率谱特征,再使用核方法SVM进行特征分类,从而实现故障诊断.经过柴油机实机测试,该方法对于不同负荷下的故障识别率达到96.72％,能有效识别不同负荷下的柴油机进排气阀间隙增大故障.
For many fault diagnosis methods based on vibration signals,the diagnosis result is usually not comprehensive under different loads.A fault diagnosis method based on power spectral density(PSD)and support vector machine(SVM)was proposed,in which the vibration signal power was processed by a moving average filter(MAF),and the PSD of the normalized signal in each period of the sample was calculated,and then the kernel method SVM was used for feature classification,so as to realize fault diagnosis.After the actual diesel engine test,the fault recognition rate of the method under different loads reaches 96.72％,and it can effectively identify the faults of the intake and exhaust valve gap increase of a diesel engine under different loads.
For many fault diagnosis methods based on vibration signals,the diagnosis result is usually not comprehensive under different loads.A fault diagnosis method based on power spectral density(PSD)and support vector machine(SVM)was proposed,in which the vibration signal power was processed by a moving average filter(MAF),and the PSD of the normalized signal in each period of the sample was calculated,and then the kernel method SVM was used for feature classification,so as to realize fault diagnosis.After the actual diesel engine test,the fault recognition rate of the method under different loads reaches 96.72％,and it can effectively identify the faults of the intake and exhaust valve gap increase of a diesel engine under different loads.