基于多重重叠技术的振动信号频谱分析方法研究

发布时间：2018-04-03 12:03

  本文选题：多重重叠技术　切入点：相对最大值抽取　出处：《振动与冲击》2017年16期

【摘要】：由于具有较小的旁瓣,汉宁、汉明和三角窗常被用于振动信号频谱分析中。但是,加这些窗会导致落在窗边缘的脉冲信号受到抑制而无法被有效检测出来,而脉冲信号中可能含有设备运行重要信息。针对此问题,研究了基于多重重叠技术的振动信号频谱分析方法并提出了相对最大值抽取的多重重叠算法。通过虚拟仿真信号和MFS-MG机械故障仿真试验台实测信号对多重重叠频谱分析方法进行了试验测试。虚拟仿真结果表明:边缘脉冲信号的多重重叠频谱幅值为一般频谱分析方法的8~10倍;试验台故障模拟信号的分析结果同样表明,多重重叠频谱分析方法比一般频谱分析方法更能表现边缘脉冲信号的宽频特征。
[Abstract]:Because of its small sidelobe, Hanning, hamming and triangular windows are often used in the spectrum analysis of vibration signals.However, adding these windows will result in the suppression of the pulse signal which falls on the edge of the window and can not be effectively detected, and the pulse signal may contain important information about the operation of the device.In order to solve this problem, the spectrum analysis method of vibration signals based on multiple overlaps is studied and a multi-overlapping algorithm for extracting the relative maximum value is proposed.The method of multiple overlapping spectrum analysis is tested by virtual simulation signal and measured signal of MFS-MG mechanical fault simulator.The virtual simulation results show that the amplitude of multiple overlapped spectrum of edge pulse signal is 8 ~ 10 times that of the general spectrum analysis method, and the analysis result of fault analog signal of the test rig also shows that,Multiplex overlapping spectrum analysis method is better than the conventional spectrum analysis method to represent the wide band characteristics of edge pulse signal.
【作者单位】： 河南工业大学机电工程学院;重庆大学机械工程学院;
【基金】：国家自然科学基金项目资助(U1604134;51541508) 河南省科技项目资助(152300410201)
【分类号】：TH17
，

本文编号：1705121