【摘要】 针对强噪声背景下的信号难于检测的问题,利用混沌阵子对周期信号的敏感性和对噪声的免疫力,提出了一种用Duffing阵子结合欧氏距离检测微弱信号的新方法。该方法采用欧氏距离确定混沌振子由混沌状态向大尺度周期状态转换的临界阈值,并利用欧氏距离的跃变自动识别混沌振子的状态。应用该方法对仿真信号和深沟球轴承的早期模拟故障信号进行分析,验证了其可行性,同时分析了相位差和噪声对于系统特性的影响,并用Simulink进行了仿真。更多还原

【Abstract】 The useful signals are often buried in heavy noise and difficult to be detected.Because chaotic oscillator is sensitive to periodic signals and immune to noise,a new weak signal detection method is proposed based on the Duffing oscillator and the Euclidean distance.The Euclidean distance is adopted to ascertain the threshold limit value of the chaotic oscillator from the chaotic status to the large-scale cycle status,and the status of chaotic oscillator is automatically identified by using jump of Euclidean distance.This method is successfully applied to the simulation signal and the incipient analog fault signal of deep groove ball bearings,and the reliability is verified.The effects of phase difference and noise on system features are analyzed,and simulation is conducted by using Simulink.更多还原