Emprical mode decomposition(EMD) is a method and principle of decomposing signal dealing with Hilbert-Huang transform (HHT) in signal analysis, while directly-mean EMD is an improved EMD method presented by N.E.Huang,...Emprical mode decomposition(EMD) is a method and principle of decomposing signal dealing with Hilbert-Huang transform (HHT) in signal analysis, while directly-mean EMD is an improved EMD method presented by N.E.Huang, the inventor of HHT, which is aimed at solving the problems of EMD principle. Although the directly-mean HMD method is very remarkable with its advantages and N. E. Huang has given a method to realize it, he did not find the theoretic evidence of the method so that the feasibility of the idea and correctness of realizing the directly-mean EMD method is still indeterminate. For this a deep research on the forming process of complex signal is made and the involved stationary point principle and asymptotic stationary point principle are demonstrated, thus some theoretic evidences and the correct realizing way of directly-mean EMD method is firstly presented. Some simulation examples for demonstrating the idea presented are given.展开更多
By virtue of neural network, a series of signals is extended forward and backward, as a result, two additional maxima and two additional minima are obtained at both ends of the original data set, with which the EMD de...By virtue of neural network, a series of signals is extended forward and backward, as a result, two additional maxima and two additional minima are obtained at both ends of the original data set, with which the EMD decomposition can be exactly achieved with cubic spline interpolation. Meanwhile, by using of neural network every IMF component can also be extended forward and backward, which effectively restrains the end effect, thus the veracious Hilbert spectra are achieved. Verifications of the sample signals and the actual surface elevation of sea waves show that the present extension method is relatively accurate.展开更多
针对强噪声背景下机械故障信号难以检测,参数辨识难度高的问题,提出了基于级联随机共振和经验模态分解的联合参数辨识方法。该方法利用EMD分层分解的思想,可以结合标准平均方差(Normalized Mean Squared Error,NMSE)准则筛选出最优IMF分...针对强噪声背景下机械故障信号难以检测,参数辨识难度高的问题,提出了基于级联随机共振和经验模态分解的联合参数辨识方法。该方法利用EMD分层分解的思想,可以结合标准平均方差(Normalized Mean Squared Error,NMSE)准则筛选出最优IMF分量,最终实现原信号频率特征参数的准确拟合。实验结果表明,文中算法可有效消除随机共振处理后信号的边缘脉冲,进而实现信号频率的准确检测。在信噪比低于-15 dB时,算法的检测性能提升了约一个数量级,在固定检测差错概率为10^(-3)时,算法的信噪比增益可达到8 dB。新算法对于机械故障信号中的频率参数辨识具有检测误差小、适应范围广泛的优势,在保证带来一定信噪比增益的同时,可实现工程器件状态的准确判断,对于提取机械系统的故障特征、识别故障类型以及进一步地排故检修具有重要意义。展开更多
基金This project is supported by National Natural Science Foundation of China(No.50275154).
文摘Emprical mode decomposition(EMD) is a method and principle of decomposing signal dealing with Hilbert-Huang transform (HHT) in signal analysis, while directly-mean EMD is an improved EMD method presented by N.E.Huang, the inventor of HHT, which is aimed at solving the problems of EMD principle. Although the directly-mean HMD method is very remarkable with its advantages and N. E. Huang has given a method to realize it, he did not find the theoretic evidence of the method so that the feasibility of the idea and correctness of realizing the directly-mean EMD method is still indeterminate. For this a deep research on the forming process of complex signal is made and the involved stationary point principle and asymptotic stationary point principle are demonstrated, thus some theoretic evidences and the correct realizing way of directly-mean EMD method is firstly presented. Some simulation examples for demonstrating the idea presented are given.
基金the NationalNatural Science Foundation of China (Grant No. 49976003) and the "863" Project (Grant No. 863-818-06-02).
文摘By virtue of neural network, a series of signals is extended forward and backward, as a result, two additional maxima and two additional minima are obtained at both ends of the original data set, with which the EMD decomposition can be exactly achieved with cubic spline interpolation. Meanwhile, by using of neural network every IMF component can also be extended forward and backward, which effectively restrains the end effect, thus the veracious Hilbert spectra are achieved. Verifications of the sample signals and the actual surface elevation of sea waves show that the present extension method is relatively accurate.
