基于多域空间状态特征的高端装备运行可靠性评价

机械设备运行可靠性对设备状态监测及故障诊断具有重要意义。传统可靠性评估方法依赖于大量故障样本,运用在单台设备可靠性评估上的实际意义有限。本文提出一种基于设备状态特征空间的运行可靠性评价方法。采集设备运行过程中的振动信号,获取时频域信息特征;采用小波包分解提取能量分布特征,构建高维特征空间。应用流形学习优化算法进行降维,获得其低维敏感特征空间,计算当前状态与正常状态的特征子空间的夹角,建立其映射关系,表征设备的当前运行可靠性。将该方法应用于不同状态下的转子实验台和滚动轴承实验台振动数据,实验结果表明:该方法对设备进行运行可靠性评价合理有效,具有很好的工程应用价值。

High-end equipment operating reliability assessment based on multi domain status space

Machinery equipment operating reliability has great significance for equipment condition monitoring and fault diagnosis. Traditional reliability evaluation method relies on large number of fault samples, and has limited practical significance when it is used on single equipment reliability evaluation. An operating reliability assessment method based on equipment status feature space is put forward. In the method, the vibration signal in the equipment operating process is acquired, from which the time and frequency domain features are extracted. The wavelet packet decomposition is adopted to extract the energy distribution features, and high dimensional feature space is established. The manifold learning optimization algorithm is applied to conduct dimension reduction, and the low dimensional sensitive feature space is obtained. The feature subspace angle between current state and normal state is calculated, and the mapping relationship is established, from which current operating reliability of the equipment is obtained. The proposed method was applied on the vibration data of rotor test rig and rolling bearing test bench under different states. The experiment results show that this method can effectively and reasonably evaluate the operating reliability of the equipment and has good engineering application value.