改进EMD阈值小波滤波方法
发布时间：2018-08-25 18:12
【摘要】：下肢自主康复训练机器人中交流伺服电机电流信号噪声严重影响电机力矩辨识精度。为解决非线性非平稳信号的滤波去噪问题,提出一种基于经验模态分解(EMD)的改进阈值小波滤波算法。首先对EMD最佳去噪层数和阈值小波的阈值处理函数进行分析和改进,然后将两种改进方法相结合,最后对Matlab中的Heavy sine信号添加高斯噪声,分别利用改进方法和软、硬阈值等滤波方法进行去噪实验。仿真实验结果表明,改进算法能有效去除非线性非平稳信号中噪声信号。与EMD和阈值小波等其他滤波方法相比,本文滤波算法去噪后信噪比更大,均方根误差更小,滤波效果更好。
[Abstract]:The motor torque identification accuracy is seriously affected by the current signal noise of AC servo motor in the autonomous rehabilitation training robot of lower extremity. An improved threshold wavelet filtering algorithm based on empirical mode decomposition (EMD) is proposed to solve the problem of filtering and denoising nonlinear non-stationary signals. First, the optimal denoising layer number of EMD and the threshold processing function of threshold wavelet are analyzed and improved, then the two improved methods are combined. Finally, Gao Si noise is added to the Heavy sine signal in Matlab, and the improved method and soft are used respectively. The denoising experiment is carried out by hard threshold and other filtering methods. Simulation results show that the improved algorithm can effectively remove noise signals from nonlinear non-stationary signals. Compared with other filtering methods, such as EMD and threshold wavelet, the SNR of this filtering algorithm is larger, the root mean square error is smaller, and the filtering effect is better.
【作者单位】： 西安交通大学机械工程学院;
【基金】：国家自然科学基金重大研究计划项目(91420301)资助
【分类号】：TN713;TP242
本文编号：2203694
[Abstract]:The motor torque identification accuracy is seriously affected by the current signal noise of AC servo motor in the autonomous rehabilitation training robot of lower extremity. An improved threshold wavelet filtering algorithm based on empirical mode decomposition (EMD) is proposed to solve the problem of filtering and denoising nonlinear non-stationary signals. First, the optimal denoising layer number of EMD and the threshold processing function of threshold wavelet are analyzed and improved, then the two improved methods are combined. Finally, Gao Si noise is added to the Heavy sine signal in Matlab, and the improved method and soft are used respectively. The denoising experiment is carried out by hard threshold and other filtering methods. Simulation results show that the improved algorithm can effectively remove noise signals from nonlinear non-stationary signals. Compared with other filtering methods, such as EMD and threshold wavelet, the SNR of this filtering algorithm is larger, the root mean square error is smaller, and the filtering effect is better.
【作者单位】： 西安交通大学机械工程学院;
【基金】：国家自然科学基金重大研究计划项目(91420301)资助
【分类号】：TN713;TP242
【相似文献】
相关期刊论文 前9条
1 潘泉;孟晋丽;张磊;程咏梅;张洪才;;小波滤波方法及应用[J];电子与信息学报;2007年01期
2 任波,李环;数字图像的一种小波滤波算法[J];沈阳工业学院学报;2004年01期
3 熊科;;小波滤波中的阈值研究[J];电力自动化设备;2008年10期
4 李顺江;贾耀曾;崔凤奎;;动不平衡信号的滤波方法研究[J];软件;2011年12期
5 吕常波;陈晓曾;翟永波;;小波滤波在光纤陀螺信号处理中的应用[J];中国西部科技;2010年06期
6 吴朝霞,吴飞,蒋学英,李志全;基于光学小波滤波的光纤光栅传感解调方法的研究[J];传感技术学报;2004年03期
7 夏洪瑞;葛川庆;董江伟;;谐波小波滤波功效分析[J];勘探地球物理进展;2007年03期
8 赵磊;赵连玉;;Haar小波滤波算法的研究[J];天津理工大学学报;2009年02期
9 孟晋丽;潘泉;张洪才;;基于相邻尺度积系数的半软阈值小波滤波[J];电子与信息学报;2007年07期
相关会议论文 前1条
1 田颜锋;刘晓刚;张冬林;郭占锋;;航空重力数据小波滤波中的频泄现象分析[A];中国地球物理·2009[C];2009年
，本文编号：2203694
本文链接：https://www.wllwen.com/kejilunwen/dianzigongchenglunwen/2203694.html
