摘要
传统的轴承故障需要大量的专业知识和复杂的特征提取工程,给故障识别技术带来了很大的问题。为了提高轴承故障诊断的性能,将深度学习算法与轴承故障诊断技术相结合,提出了一种基于融合卷积神经网络(CNN)和海洋捕食者(MPA)算法优化支持向量机(SVM)的滚动轴承故障诊断方法。首先,将原始时域振动信号分为一维和二维2种形式,分别输入一维和二维卷积神经网络进行自适应特征提取;其次,在聚合层进行特征信息融合;最后,利用最新的优化算法MPA对SVM进行优化。实验结果表明:经过MPA优化后的轴承故障诊断算法的准确率最高为99.6%,说明本算法具有一定的研究价值。
The traditional bearing fault requires a lot of professional knowledge and complex feature extraction engineering,which brings big problems to the fault recognition technology.In order to improve the performance of bearing fault diagnosis,a rolling bearing fault diagnosis method based on a convolution neural network(CNN)and marine predator algorithm(MPA)optimized support vector machine(SVM)is proposed by combining deep learning algorithm with bearing fault diagnosis technology.Firstly,the original time-domain vibration signal is divided into one-dimensional and twodimensional forms,which are inputs of one-dimensional and two-dimensional convolutional neural network for adaptive feature extraction.Then feature information fusion is carried out in the aggregation layer.Finally,the latest optimization algorithm MPA is used to optimize SVM.The experiment result shows that the accuracy of the bearing fault diagnosis algorithm optimized by MPA is up to 99.6%,which proves that this algorithm has certain research value.
作者
林珍莉
曾宪文(指导)
LIN Zhenli;ZENG Xianwen(School of Electronic Information Engineering,Shanghai Dianji University,Shanghai 201306,China)
出处
《上海电机学院学报》
2022年第5期254-259,共6页
Journal of Shanghai Dianji University
