欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
为了改进经验模式分解（EMD）算法的消噪性能,在传统EMD消噪分解的基础上,参照小波阈值的消噪方法,提出了一种基于自适应阈值的EMD消噪方法.首先,建立去噪阈值和均方误差之间的对应函数,在所选阈值保证均方误差最小的前提下,利用具有较好全局搜索性的蚁群算法,根据建立的函数搜索阈值,克服了传统方法中硬阈值和软阈值固定选取的缺陷,实现了最优阈值的选取.仿真信号分析和实际轴承故障信号分析表明,该方法与传统的EMD消噪方法、软硬阈值分析方法相比,消噪效果更加明显.
英文摘要：
To further enhance the de--noising performance of EMD algorithm, this paper proposed a de-noising method based on adaptive threshold of EMD referred to the method of threshold-based wavelet de-noising. First, the relationship between the mean square error（MSE） and the threshold function were established,then the optimal threshold of each intrinsic mode function（IMF） level were searched and obtained by the ant colony algorithm which realized to get the minimum MSE. So this method eliminates the disadvantages of soft and hard threshold de-noising method effectively. The experi- mental results indicate that the method is more effective in noise reduction in comparison with the convention- al soft and hard threshold-based EMD de-noising methods, the method works better.
同期刊论文项目
同项目期刊论文
期刊信息
位置：