摘要
在295柴油机上进行设定的进排气系统故障实验,获取各工况下三个不同测点位置的缸盖振动信号,然后利用改进的基于神经网络和小波分析的故障诊断方法进行分析.实验和仿真结果表明,不同测点获取的振动信号蕴含故障特征是不同的.从缸盖上方采集的振动信号更能表征各种故障特征,其故障准确识别率达到95.4%,较其他两个测点有大幅度提高.为基于振动信号的柴油机故障诊断提供了最佳的测点位置,以提高故障的诊断准确性.对其他复杂机械的振动诊断同样具有参考价值.
Experiments of the intake and exhaust system with designed faults are set to measure the vibration signals in the cylinder head from three different directories. The improved diagnosis method based on neural network and wavelet theory was used to analysis them. According to the experiment and simulation result, the fault characteristics contained in the vibration signals from three measure locations are different. The identification veracity of the signals measured upon the cylinder head is 95.4 percent. It is better than others. So the best location can be chosen to measure the vibration to improve the fault diagnosis veracity. And it has the referenced value to other vibration diagnosis of complex machines.
出处
《华中科技大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2007年第5期99-101,共3页
Journal of Huazhong University of Science and Technology(Natural Science Edition)
关键词
柴油机
故障诊断
神经网络
小波分析
diesel engine
fault diagnosis
neural network
wavelet analyses
