提出了基于局部均值分解(Local mean decomposition,简称LMD)和AR模型相结合的转子系统故障诊断方法.该方法先用LMD方法将转子振动信号分解成若干个瞬时频率具有物理意义的PF(Product function,简称PF)分量之和,然后对每一个PF分量建立A...提出了基于局部均值分解(Local mean decomposition,简称LMD)和AR模型相结合的转子系统故障诊断方法.该方法先用LMD方法将转子振动信号分解成若干个瞬时频率具有物理意义的PF(Product function,简称PF)分量之和,然后对每一个PF分量建立AR模型,提取模型参数和残差方差作为故障特征向量,并以此作为神经网络分类器的输入来识别转子的工作状态和故障类型.与EMD方法的对比研究表明,这两种方法均能有效地应用于转子系统的故障诊断.但LMD方法信号分解后数据残差比EMD方法的小.展开更多
Ionospheric peak value of F2 layer(Nm F2)is an important parameter in the ionosphere,which has important applications in short-wave communication,ionospheric modeling and so on.In this paper,the empirical orthogonal f...Ionospheric peak value of F2 layer(Nm F2)is an important parameter in the ionosphere,which has important applications in short-wave communication,ionospheric modeling and so on.In this paper,the empirical orthogonal function(EOF)decomposition method is used to analyze the Nm F2obtained from the occultation data.Daily spatial distribution of Nm F2at the same time is relatively even.Variance of first modal is much larger than the other modals.A local wavelet power spectrum(LWPS)method is applied to analysis the cycle of F10.7index and time coefficient of first modal.The result shows that they have similar cycle distribution,indicating that F10.7index is the main factor affecting variation of Nm F2.A function is established between the tine coefficient of first modal and F10.7index,average F10.7index value of early 81 days fp by least squares method.The results show that contribution coefficient of fp is negative which indicates that fp has an inert effect existing in the ionosphere.Contribution coefficient of F10.7is positive,which is consistent with the fact that it has an anomaly in winter/spring seasons.In summary,it is feasible to establish a mid-latitude empirical Nm F2model in northern hemisphere based on occultation data and EOF decomposition method.展开更多
文摘提出了基于局部均值分解(Local mean decomposition,简称LMD)和AR模型相结合的转子系统故障诊断方法.该方法先用LMD方法将转子振动信号分解成若干个瞬时频率具有物理意义的PF(Product function,简称PF)分量之和,然后对每一个PF分量建立AR模型,提取模型参数和残差方差作为故障特征向量,并以此作为神经网络分类器的输入来识别转子的工作状态和故障类型.与EMD方法的对比研究表明,这两种方法均能有效地应用于转子系统的故障诊断.但LMD方法信号分解后数据残差比EMD方法的小.
基金supported by the National Natural Science Foundation of China(Grant No.40505005)the Specialized Research Fund for State Key Laboratories(Grant No.Y22612A33S)
文摘Ionospheric peak value of F2 layer(Nm F2)is an important parameter in the ionosphere,which has important applications in short-wave communication,ionospheric modeling and so on.In this paper,the empirical orthogonal function(EOF)decomposition method is used to analyze the Nm F2obtained from the occultation data.Daily spatial distribution of Nm F2at the same time is relatively even.Variance of first modal is much larger than the other modals.A local wavelet power spectrum(LWPS)method is applied to analysis the cycle of F10.7index and time coefficient of first modal.The result shows that they have similar cycle distribution,indicating that F10.7index is the main factor affecting variation of Nm F2.A function is established between the tine coefficient of first modal and F10.7index,average F10.7index value of early 81 days fp by least squares method.The results show that contribution coefficient of fp is negative which indicates that fp has an inert effect existing in the ionosphere.Contribution coefficient of F10.7is positive,which is consistent with the fact that it has an anomaly in winter/spring seasons.In summary,it is feasible to establish a mid-latitude empirical Nm F2model in northern hemisphere based on occultation data and EOF decomposition method.
基金supported by the National Natural Science Foundation of China(Nos.51079032,51490671,and 11572093)the International Science and Cooperation Sponsored by the National Ministry of Science and Technology of China(No.2012DFA70420)
