斜拉索振动信号的包络线稀疏复原算法

发布时间：2018-05-26 03:04

本文选题：经验模态分解 + Hilbert变换　；参考：《振动与冲击》2015年23期

【摘要】：经验模态分解算法在故障诊断、信号去噪、趋势预测和趋势消除等很多领域具有广泛的应用价值。信号包络线提取是经验模态分解算法的核心关键技术,直接影响分解结果的效果。目前在信号处理中常用的包络分析法有Hilbert变换、广义检波滤波、三次样条插值法和偏微分方程建模等,但是这些方法存在提取包络线精度不高、端点效应等不足,尤其是端点抖动效应导致很大的包络线提取误差。将信号的极值点看成是包络线信号的某一变换域上稀疏采样点,引入了稀疏优化算法对这些极值点进行包络线的稀疏优化复原。首先研究分析包络线的平稳变化特性,以此构建变频的DCT稀疏基;其次求解信号的极值点,以信号的极值点集用于稀疏优化算法的观测值;然后采用正交匹配追踪算法进行包络线的稀疏复原求解;最后对实际斜拉索振动信号进行算法测试和应用,通过与三次样条插值法进行比较分析,结果表明本算法不仅可以提高提取信号包络线的精度,还可以有效的抑制端点效应。
[Abstract]:Empirical mode decomposition (EMD) algorithm has been widely used in many fields such as fault diagnosis, signal denoising, trend prediction and trend elimination. Signal envelope extraction is the key technology of empirical mode decomposition (EMD) algorithm, which directly affects the result of decomposition. At present, the envelope analysis methods commonly used in signal processing include Hilbert transform, generalized demodulation filtering, cubic spline interpolation and partial differential equation modeling, etc. In particular, the end point jitter effect leads to a large envelope extraction error. The extremum of the signal is regarded as a sparse sampling point in a transform domain of the envelope signal, and a sparse optimization algorithm is introduced to restore these extreme points to the sparse optimization of the envelope. Firstly, the stationary variation of the envelope is analyzed to construct the frequency conversion DCT sparse basis, and then the extreme point of the signal is solved, and the set of the extreme points of the signal is used for the observation value of the sparse optimization algorithm. Then we use the orthogonal matching tracking algorithm to solve the sparse restoration of the envelope. Finally, we test and apply the actual cable vibration signal, and compare it with the cubic spline interpolation method. The results show that the proposed algorithm can not only improve the accuracy of signal envelope extraction, but also suppress the endpoint effect effectively.
【作者单位】：宁波大学信息科学与工程学院;
【基金】：国家自然科学基金(61071198) 浙江省自然科学基金(LY13F010015) 浙江省重点科技创新团队子项目(421400250) 浙江省重中之重学科开放基金(xkx11417) 宁波市自然科学基金(2012A610019)
【分类号】：U441.3;U448.27

【参考文献】