基于声振信号和Transformer的圆锥滚子轴承故障诊断方法研究  

Research on Tapered Roller Bearing Fault Diagnosis Method Based on Acoustic Vibration Signal and Transformer

作　　者：王子豪 王风涛 熊元昊 胡明涛 田召阳 WANG Zihao;WANG Fengtao;XIONG Yuanhao;HU Mingtao;TIAN Zhaoyang(School of Mechanical and Automotive Engineering,Anhui Polytechnic University,Wuhu 241000,China)

机构地区：[1]安徽工程大学机械与汽车工程学院,安徽芜湖241000

出　　处：《东莞理工学院学报》2025年第1期114-121,共8页Journal of Dongguan University of Technology

基　　金：国家自然科学基金项目(51905001);安徽未来技术研究院企业合作项目(2023qyhz22);安徽工程大学校级项目(Xjky2022012)。

摘　　要：轴承的故障监测工作中,受使用环境影响,单一振动难以获取,声音信号成为一种有效选择。提出一种基于声振信号和Transformer相结合的圆锥滚子轴承故障诊断方法。该方法以声音信号代替振动信号作为神经网络的输入,之后使用Transformer网络对轴承正常信号和故障信号进行分类,从而精确判断轴承故障类型,所获得的分析结果通过实验进行验证。实验证明:Transformer不仅有着98.64%准确率而且运行速度更快。这不仅验证了基于声音信号进行轴承故障诊断的可行性,还解决了在振动信号难以获得时的轴承故障诊断工作,体现出声音在轴承故障诊断中的应用价值。In the fault monitoring work of bearings,a single vibration is difficult to obtain due to the influence of the use environment,and the sound signal becomes an effective choice.This paper proposes a tapered roller bearing fault diagnosis method based on the combination of sound and vibration signals and Transformer.The method takes sound signal instead of vibration signal as the input of the neural network,and then uses Transformer network to classify the normal and faulty signals of the bearing,so as to accurately determine the type of bearing faults,and the analysis results obtained are verified by experiments.Experiments shows that Transformer not only has 98.64%accuracy but also runs faster.This not only verifies the feasibility of bearing fault diagnosis based on voice signal,but also solves the bearing fault diagnosis when the vibration signal is difficult to obtain,and reflects the application value of voice in bearing fault diagnosis.

关 键 词：声音 TRANSFORMER 圆锥滚子轴承 故障诊断 

分 类 号：TH133.33[机械工程—机械制造及自动化]