噪声协助的EMD-1.5维谱信号抗混分解与特征提取

针对大型动力装备核心部件微弱故障特征信息提取问题，提出了一种噪声协助的EMD-1.5维谱故障诊断方法。经验模式分解(EMD)方法中，信号极值点间隔特性影响模式混淆现象的出现，针对此状况提出信号极值点间隔特性评价方法，分析高斯白噪声有助于信号抗混分解原理，通过对原始信号加入高斯白噪声得到噪声协助的EMD方法，提高信号抗混分解能力。将1.5维谱与噪声协助的EMD方法结合，得到一种新的故障特征提取方法，该方法具有对信号进行有效抗混分解、提取非线性耦合等特性，并以此来提取故障的微弱特征信息。通过仿真研究与电力机车滚动轴承的故障诊断工程实例，验证了该方法的有效性。

Noise Assisted EMD-1.5 Dimension Spectrum for Signal Anti Alias Decomposition and Feature Extraction

To effectively extract the weak fault features of core components of the large-scale power equipment, a new fault diagnosis method based on noise assisted EMD- 1.5 dimension spectrum is proposed. The characteristic of signal extreme intervals affects the appearance of mode mixing on EMD method, according to this status, the evaluation method of the characteristic of signal extreme intervals is proposed, the principle that Gaussian white noise helping to avoid mode mixing is analyzed. Noise assisted EMD method is get by adding Gaussian white to original signal to improve the capability of signal anti alias decomposition. Meanwhile, noise assisted EMD combines with 1.5 dimension spectrum to get a new fault diagnosis method, which has the capability of effectively anti alias decomposition and extracting the nonlinear coupling feature, to extract the weak fault features. Finally, noise assisted EMD- 1.5 dimension spectrum is effectively verified by simulation experiment and engineering example of electric locomotive rolling bearing fault diagnosis.