中文摘要：

提出了一种Hilbert-Huang变换（HHT）和神经网络相结合的智能轴承状态监控系统.从理论上阐述了经验模态分解（EMD）、固有模态函数（IMF）、Hilbert变换以及广义回归神经网络（GRNN）.提出了轴承智能状态监控流程图即小波包对振动信号进行去噪预处理,HHT提取IMF的瞬时幅值作为特征向量,实验使用了BPNN和GRNN 2种神经网络,通过神经网络进行故障辨识和分类,最后用轴承的振动数据对该系统进行验证.结果表明,提出的状态监控系统能较好地对轴承的状态进行监控.

英文摘要：

An intelligent bearing condition monitoring system using Hilbert-Huang transform（HHT） and artificial neural network was proposed.The theory of empirical mode decomposition（EMD）,intrinsic mode functions（IMF） and the Hilbert transform were introduced,and generalized regression neural network（GRNN） theory also described,then flow chart of intelligent condition monitoring system for bearing was proposed,wavelet packet was used in the process of denoising pretreatment for vibration signal,HHT extracted the instantaneous amplitude of the IMF as feature vectors,and the neural network was taken for fault identification and classification.Finally,the system was verified by the vibration data.And both GRNN and BPNN were used and compared in the experiment platform.Experimental results indicate that the proposed condition monitoring system can distinguish the bearing condition better.