基于机械振动信号分析的轴承故障分类方法  

Bearing Fault Classification Method Based on Mechanical Vibration Signal Analysis

作　　者：王建国 张玄 陈跃 梁樱紫 WANG Jianguo;ZHANG Xuan;CHEN Yue;LIANG Yingzi(Yankuang Energy Group Company Limited,Jining 273500,China;Shandong Institutes of Industrial Technology,Jinan 250000,China;Shandong Industrial Technology Research Institute Bozheng Innovation Consulting Co.,Ltd.,Jinan 250000,China;Jinan Shengquan Group Share Holding Co.,Ltd.,Jinan 250000,China)

机构地区：[1]兖矿集团,山东济宁273500 [2]山东产业技术研究院,济南250000 [3]山东产研博正创新咨询有限公司,济南250000 [4]济南圣泉集团股份有限公司,济南250000

出　　处：《智能物联技术》2024年第6期47-50,共4页Technology of Io T& AI

摘　　要：提出一种基于机械振动信号分析的轴承故障分类方法。首先,通过对机械振动信号进行短时傅里叶变换(Short-Time Fourier Transform,STFT),提取时频域特征向量;其次,利用自注意力机制增强的多层感知机(Multi-LayerPerceptron,MLP)模型进行轴承故障分类,该模型能够有效捕捉特征间的依赖关系,从而提高分类精度;最后,采用PRONOSTIA轴承数据集进行模型训练和测试。实验结果表明,该方法在正常状态和磨损故障的识别上具有出色的表现,尽管裂纹故障的检测精度略逊一筹,但总体准确率仍达到0.93。This paper presents a bearing fault classification method based on mechanical vibration signal analysis.Firstly,the feature vector in time-frequency domain is extracted by Short-Time Fourier Transform(STFT)on the mechanical vibration signal.Secondly,the Multi-Layer Perceptron(MLP)model enhanced by self-attention mechanism is used to classify bearing faults,which can effectively capture the dependency between features to improve the classification accuracy.Finally,PRONOSTIA bearing data set was used for model training and testing.The experimental results show that this method has excellent performance in the identification of normal state and wear faults.Although the detection accuracy of crack faults is slightly lower,the overall accuracy is still 0.93.

关 键 词：机械信号 故障分类 短时傅里叶变换(STFT) 自注意力机制 多层感知机(MLP) 

分 类 号：TP311.5[自动化与计算机技术—计算机软件与理论]