Abstract:The end effect of empiricd mode decomposition (EMD)that appears when using spline interpolation to get two envelops of the data is discussed. The shortage of ordinary end extending methods is analyzed. An extending method based on lyapunov exponent forecasting mode is proposed. Analyzing the shortages and problems of Lyapunov exponent forecasting model, then a new forecasting model based on local radiation degrees which is similar to Lyapunov exponent forecasting model is proposed. By using the improved lyapunov exponent forecasting mode, the data sequence is extended by adding some predicted points, and it is stopped when a new local maximum and minimum points are emerged. Then the EMD could be decomposed exactly. By comparing the proposed method with the other ordinary end extending methods, it can be find that the proposed method is effective when dealing with the end effect. In the end, the proposed method is used in the fault diagnosis of rotary machine, the result demonstrates that end effect can be controled effectively and faults can be recognized exactly by using the proposed method.
蔡艳平;李艾华;石林锁;何艳萍;赵静茹. EMD端点效应的改进型混沌延拓方法及其在机械故障诊断中的应用[J]. , 2011, 30(11): 46-52.
Cai Yan-ping;Li Ai-hua;Shi Lin-suo;He Yan-ping;Zhao Jing-ru. Processing method for end effects of emd based on improved chaos forecasting model and its application in machinery fault diagnosis. , 2011, 30(11): 46-52.