摘要
往复机在发生故障时会使得正常状态下的振动频带有所改变,因此各频率成分的能量分布中包含有丰富的故障信息,可以作为设备状态的评价参数。文中提出了基于"频带能量分布-设备状态"的故障辨识方法,采用小波包进行信号的分解-重构,并对每个频段的能量进行归一化,构成一个特征向量,可以以此进行设备的状态识别,并以柴油机的状态识别证明了此方法的有效性。
Vibration frequency-band will be different from normal condition when reciprocating engine is fault condition. Therefore, the energy of different frequency-band contains more information about reciprocating engine working condition. It can be used the parameters to evaluate reciprocating engine working condition. In this paper, a new condition pattern recognition method has been presented which is named as ＂energy frequency band-engine working condition＂. Wavelet packet has been used for vibration signal decomposition and reconstruction. Then, energy distribution of every frequency band can be calculated according to normalization process. A new feature vector can be obtained to be used for pattern recognition. Diesel engine pattern recognition process is used as an example to testify the effectiveness of this method.
出处
《机械工程师》
2008年第3期35-37,共3页
Mechanical Engineer
基金
国家自然科学基金项目(50475155)
关键词
小波包
频带能量
状态识别
wavelet-packet
frequency-band-energy
pattern recognition
