摘要
为了更高效率地诊断轴承故障,提出了一种基于精英反向学习(OBL)改进麻雀算法(SSA)优化卷积神经网络(CNN)的滚动轴承故障诊断方案,利用改进SSA较强的寻优能力和较快的收敛速度,优化CNN的超参数。以美国凯斯西储大学的滚动轴承数据作为实验数据对该方案加以检验,并与BP神经网络、支持向量机(SVM)以及未优化的CNN模型等故障诊断方式相比较。结果表明:该方案分类准确度更高,用时更少。
In order to diagnose bearing faults more efficiently, a rolling bearing fault diagnosis scheme is proposed based on a convolutional neural network(CNN), which is optimized by the improved sparrow search algorithm(SSA) based on elite opposition-based learning(OBL). The strong optimization ability and fast convergence speed of the improved SSA are used to optimize the hyperparameters of CNN. The method is tested on the experimental data, which is the rolling bearing data from Case Western Reserve University, and compared with some fault diagnosis methods such as BP neural network, support vector machine(SVM) and unoptimized CNN model. The results show that the method has higher classification accuracy and less time.
作者
程渠超
刘湲
CHENG Quchao;LIU Yuan(School of Electrical Engineering,Shanghai Dianji University,Shanghai 201306,China)
出处
《上海电机学院学报》
2022年第1期40-45,共6页
Journal of Shanghai Dianji University