应用混合2D-CNN-LSTM模型诊断轴承故障OA
Application of Hybrid 2D-CNN-LSTM Model for Diagnosing Bearing Faults
滚动轴承是机械设备中的核心组件.为保证滚动轴承故障诊断的准确性和实时性,本文提出了一种混合深度模型故障诊断方法.该方法结合二维卷积神经网络和长短期记忆网络,不仅提升了模型的性能,而且能准确捕捉轴承信号中的空间和时间特征.实验结果表明,该方法能够准确分类轴承故障,并实现对轴承运行状态的实时监测.
Rolling bearings are the core components in mechanical equipment.To ensure the accuracy and real-time diagnosis of rolling bearing faults,this paper proposes a hybrid deep model fault diagnosis method.This method combines two-dimensional convolutional neural networks and long short-term memory networks,not only improving the performance of the model,but also accurately capturing the spatial and temporal features in bearing signals.The experimental results show that this method can accurately classify bearing faults and achieve real-time monitoring of bearing operation status.
江跃龙
广州铁路职业技术学院信息工程学院 广州 511300
计算机与自动化
二维卷积神经网络长短期记忆网络轴承故障分类识别
2D-CNNLSTMBearing FaultClassification Recognition
《福建电脑》 2024 (005)
33-37 / 5
本文得到广东省普通高校创新团队项目(自然科学)(No.2021KCXTD068)、广州市高等教育教学质量与教学改革工程计算机应用技术专业群"双师型"教师培养培训基地(No.2022SSPRJD004)、科教融合专项工艺改进项目基于改进的深度学习网络芯片引脚及表面缺陷检测方法(No.GTXYK2208)资助.
评论