摘要
针对一些轴承传统故障诊断算法模型简易造成的诊断结果精度不高且稳定度不高的问题,进一步提高故障诊断的精度和稳定度,文章提出了基于深度学习的轴承故障智能诊断方法。首先采用小波对数据进行去噪处理;然后使用深度卷积神经网络训练数据自动学习特征,并加入Dropout正则化技术避免模型的过拟合;最后使用机器学习SKlearn库下的SVC模块搭建SVM分类器并使用Adam算法进行模型的搭建和优化。实验结果表明,文章提出的故障诊断方法平均故障诊断率达到99.7%,对工业生产中的设备故障诊断具有较大意义。
In order to improve the accuracy and stability of bearing fault diagnosis,an intelligent bearing fault diagnosis method based on deep learning is proposed. Firstly,wavelet is used to denoise the data. Then,the deep convolutional neural network is used to train the data to automatically learn the features,and Dropout regularization is added to avoid the overfitting of the model. Finally,the SVC module under the machine learning SKlearn library is used to build SVM classifier and Adam algorithm is used to build and optimize the model. The experimental results show that the average fault diagnosis rate of the proposed fault diagnosis method is 99.7%,which is of great significance to equipment fault diagnosis in industrial production.
作者
黄扣
袁伟
陈红卫
HUANG Kou;YUAN Wei;CHEN Hongwei(School of Electronic Information,Jiangsu University of Science and Technology,Zhenjiang 212100)
出处
《计算机与数字工程》
2022年第8期1827-1832,共6页
Computer & Digital Engineering
基金
国家自然基金项目(编号:71972090)资助。