变载荷齿轮箱故障信号智能检测方法

针对变载荷激励下齿轮故障信号检测和故障识别的问题,提出一种基于经验模态分解和深度信念网络的变载荷齿轮箱故障信号智能检测方法。首先通过经验模态分解方法将非平稳的加速度振动信号分解成若干平稳的本征模态函数;选出啮合频率及倍频所在的本征模态函数,重构信号,求出重构信号的频谱,作为深度信念网络的输入;深度信念网络通过对输入频谱进行预训练和特征学习,并建立变载荷激励下齿轮故障识别分类模型;最后,用构建好的深度信念网络对测试样本进行故障诊断。试验结果表明,提出的方法能有效地检测和识别变载荷激励下齿轮故障。

Intelligent Detection Method of Variable Load Gearbox Fault Signal

Aiming at the problem of gear fault feature extraction and classification under variable load excitation, an intelligent detection method of gearbox based on empirical mode decomposition (EMD) and deep belief network (DBN) is presented. Therefore, the intrinsic mode function (IMF) of the meshing frequency and the frequency doubling is selected to reconstruct the signal, and obtain the spectrum of the reconstructed signal, which is the input of the deep belief network. In the deep belief network, the pre-training and feature learning of input spectrum are carried out, and the classification model of gear fault recognition based on variable load excitation is established. Finally, fault diagnosis is carried out by using the constructed deep belief network. The experimental results show that the proposed method can effectively identify gear failure types under variable load excitation.