大型船舶动力装置故障的诊断研究
|
摘要
大型船舶动力装置工作负荷较大,故障诱因较多,为了有效准确排除大型船舶动力装置,提高工况稳定性,提出一种基于喘振谱特征提取的大型船舶动力装置故障检测诊断方法。采用多传感器进行大型船舶动力装置的物理信息采集,提取动力装置的振动信号和喘振信号,对提取的信号进行时频变换和离散谱特征分析,采用自相关匹配滤波器进行船舶动力装置振动传感信号的滤波处理,对滤波输出信号进行喘振谱提取,对提取的谱特征量输入到神经网络分类器中进行故障判别。仿真及结果表明,采用该方法进行大型船舶动力装置故障诊断的准确性较高。
The large ship power plant work load, fault inducement more effectively and accurately in order to exclude large ship power plant, improve the operation stability, a surge of spectral feature extraction of large ship power equipment fault detection and diagnosis method based on physical information acquisition using multiple sensors for large marine propulsion plant, vibration signal extraction and the surge signal power device, and the frequency characteristics of discrete spectrum transform analysis extracted signal, the filtering device for vibration sensing signal of ship dynamic autocorrelation matching filter, spectrum extraction of surge filter output signal, the spectral feature extraction of the input to the neural network classifier for fault simulation results and discrimination. Show that, using the method of fault diagnosis of large ship power device of high precision.
The large ship power plant work load, fault inducement more effectively and accurately in order to exclude large ship power plant, improve the operation stability, a surge of spectral feature extraction of large ship power equipment fault detection and diagnosis method based on physical information acquisition using multiple sensors for large marine propulsion plant, vibration signal extraction and the surge signal power device, and the frequency characteristics of discrete spectrum transform analysis extracted signal, the filtering device for vibration sensing signal of ship dynamic autocorrelation matching filter, spectrum extraction of surge filter output signal, the spectral feature extraction of the input to the neural network classifier for fault simulation results and discrimination. Show that, using the method of fault diagnosis of large ship power device of high precision.
引文
[1]敖星,吕俊峰,赵孟娜.基于主成分分析和多元支持向量机的旋转机械故障诊断方法[J].四川兵工学报,2010,31(9):83-86.
[2]高闪,梅劲松.输入非线性系统的执行器故障容错控制[J].信息与控制,2015,44(4):463-468.
[3]温阳东,宋阳,王颖鑫,等.基于模糊神经网络的电力变压器故障诊断[J].计算机测量与控制,2013,21(1):39-41.
[4]李兵,徐榕,贾春宁,等.基于自适应形态提升小波与改进非负矩阵分解的发动机故障诊断方法[J].兵工学报,2013,34(3):353-360.
[5]MAHBOUBI H.Distributed deployment algorithms for efficient coverage in a network of mobile sensors with nonidentical sensing Capabilities[J].IEEE Transactions on Vehicular Technology,2014,63(8):3998-4016.
[2]高闪,梅劲松.输入非线性系统的执行器故障容错控制[J].信息与控制,2015,44(4):463-468.
[3]温阳东,宋阳,王颖鑫,等.基于模糊神经网络的电力变压器故障诊断[J].计算机测量与控制,2013,21(1):39-41.
[4]李兵,徐榕,贾春宁,等.基于自适应形态提升小波与改进非负矩阵分解的发动机故障诊断方法[J].兵工学报,2013,34(3):353-360.
[5]MAHBOUBI H.Distributed deployment algorithms for efficient coverage in a network of mobile sensors with nonidentical sensing Capabilities[J].IEEE Transactions on Vehicular Technology,2014,63(8):3998-4016.