摘要
引入稀疏混沌时间短序列概念,提出了稀疏符号时间序列分析方法与实现流程。利用映射空间关系对混沌振子的相变进行自治而快速辨识,以尽早而准确地捕捉表征转子碰摩早期的特征信号,进而实现对转子碰摩故障进行早期检测。仿真和数值实验均表明,基于稀疏符号时间序列分析方法可以对信噪比为-37.163dB左右的微弱信号进行可靠检测,是一种具有实际应用价值的故障早期特征检测方法。
The concept of sparse symbolic time series analysis was brought in and the method and procedure for implementation of the analysis was proposed.The mapping space relationship was utilized to determine the phase transition of oscillator autonomously and quickly in order to catch the characteristic signal of mechanical faults as early and accurately as possible,and then the early detection of rub-impact fault in rotor was realized.The simulation and numerical analysis demonstrate that the method can reliably detect weak signal with signal-to-noise ratio as small as-37.163dB,and the method is feasible for feature detection of early faults such as rub-impact fault of rotor.
出处
《振动与冲击》
EI
CSCD
北大核心
2008年第12期148-150,166,共4页
Journal of Vibration and Shock
关键词
稀疏符号时间序列分析
转子系统
碰摩故障
早期检测
sparse symbolic time series analysis
rotor system
rub-impact fault
early detection
