摘要
指向李雅普诺夫指数(directional Lyapunov exponent,DLE)描述了一个系统在相空间中平衡点附近的特征结构,它与系统参数之间有着直接联系,因而可用来实现对被测系统的故障诊断。但由于对其计算时要求输出相空间轨线在平衡点附近形成单涡结构,因而对激励信号形式提出了很高的要求。利用固有时间尺度分解(intrinsic time-scale decomposition,ITD)方法,将任意混沌激励下的输出信号分解为一组单涡混沌信号,进而对各分量求解DLE曲线;并通过分析DLE曲线所反映的被测系统平衡点附近各个方向上平均发散速率大小的变化,实现对被测模拟电路系统的故障诊断。仿真结果表明,该方法可以有效地对被测系统平衡点附近结构进行分析,可应用于模拟电路的故障诊断之中。
Directional Lyapunov exponent (DLE) depicts the eigen-structure around a system's equilibrium points in the phase space, which closely associates with the parameters of the system, and it can be used to detect the fault of the system under test (SUT). But, the calculation of the DLE curves needs the output phase space trajectory to be a single roll, which limit the style of the excitation. The output of the SUT under chaotic excitations is decomposed into a group of signals that are single roll in its corresponding Hilbert space by using the intrinsic timescale decomposition (ITD). Then DLE curves are calculated for each single roll trajectory, and the parameter changes are detected by analyzing the DLE curves. Simulation results show that, the proposed methods can analyze the eigen-structure around the equilibrium point in the phase space, and it can be used for the fault diagnosis of the analog circuit system.
出处
《科学技术与工程》
北大核心
2015年第12期70-74,80,共6页
Science Technology and Engineering
基金
国家自然科学基金(61174207,11405267)资助