摘要
传统的机器学习方法在行星齿轮箱故障诊断方面存在识别率低、特征提取操作繁琐等问题。为提高行星齿轮箱的诊断效率,提出基于一维深度卷积神经网络(One-dimensional deep convolutional neural network,1-DCNN)的故障诊断方法,将原始信号直接输入到网络中进行诊断。通过对行星齿轮箱行星轮5种故障信号进行训练验证,精度可达100%,且在诊断精度和效率上优于其他常用算法。
Traditional machine learning methods have disadvantages such as low recognition rate and complicated feature extraction operations in the planetary gearbox fault diagnosis.In order to improve the diagnosis efficiency of planetary gearboxes,a fault diagnosis method based on one-dimensional deep convolutional neural network(1-DCNN)is proposed,and the original signals are directly input to the network for diagnosis.The accuracy of diagnosing five kinds of fault signals of planetary gear of planetary gear box can reach 100%,and the diagnostic accuracy and efficiency are better than other commonly used algorithms.
作者
薛璇怡
庞新宇
Xue Xuanyi;Pang Xinyu(College of Mechanical and Vehicle Engineering,Taiyuan University of Technology,Taiyuan 030024,China;Shanxi Key Laboratory of Fully Mechanized Coal Mining Equipment,Taiyuan 030024,China)
出处
《机械传动》
北大核心
2020年第11期127-133,共7页
Journal of Mechanical Transmission
基金
国家自然科学基金(51805352)
山西省自然科学基金(201901D111062)。
