基于堆栈降噪自编码网络的机械设备磨损状态识别

樊红卫; 马宁阁; 张旭辉; 高烁琪; 曹现刚; 马宏伟

doi:10.13272/j.issn.1671-251x.17633

基于堆栈降噪自编码网络的机械设备磨损状态识别

樊红卫^{1, 2},
马宁阁¹,
张旭辉^{1, 2},
高烁琪¹,
曹现刚^{1, 2},
马宏伟^{1, 2}

1.
西安科技大学机械工程学院，陕西西安710054
2.
陕西省矿山机电装备智能监测重点实验室，陕西西安710054

基金项目:

国家自然科学基金重点项目（51834006）

陕西省重点研发计划项目（2018ZDCXL-GY-06-04，2019GY-093）

陕西省自然科学基础研究计划项目（2019JLZ-08）

详细信息

中图分类号: TD67
计量
- 文章访问数: 68
- HTML全文浏览量: 16
- PDF下载量: 15
出版历程
- 刊出日期: 2020-11-19

Wear state recognition of mechanical equipment based on stacked denoised auto-encoding network

FAN Hongwei^{1, 2},
MA Ningge¹,
ZHANG Xuhui^{1, 2},
GAO Shuoqi¹,
CAO Xiangang^{1, 2},
MA Hongwei^{1, 2}

1.
College of Mechanical Engineering, Xi''an University of Science and Technology,Xi''an 710054, China
2.
Shaanxi Key Laboratory of Mine Electromechanical Equipment Intelligent Monitoring, Xi''an 710054, China)

摘要

摘要: 通过磨粒铁谱图像识别可实现机械设备磨损状态识别，但基于机器学习的磨粒铁谱图像识别存在较多人工干预、普适性差。为解决上述问题，提出了一种基于堆栈降噪自编码网络的机械设备磨损状态识别方法。将多个降噪自编码网络进行堆叠，即上一个降噪自编码网络隐含层的输出作为下一个降噪自编码网络的输入，并在最后一个降噪自编码网络隐含层后添加Softmax分类器，从而构建堆栈降噪自编码网络；利用磨粒铁谱图像对堆栈降噪自编码网络进行无监督预训练，并通过有监督微调优化网络参数，对磨粒铁谱图像进行分类，实现机械设备磨损状态智能识别。实验结果表明：当堆栈降噪自编码网络的激活函数为Relu、优化器为Adam、学习率为0.001时，网络性能最佳，识别准确率达98.43%。
- 机械设备 /
- 磨损状态识别 /
- 铁谱图像 /
- 堆栈降噪自编码网络
Abstract: Wear state recognition of mechanical equipment can be realized by image recognition of ferrography image of wear particle, but ferrography image of wear particle recognition based on machine learning has more manual intervention and poor universality. In order to solve the above problems, a wear state recognition method of mechanical equipment based on stacked denoised auto-encoding network was proposed. Multiple denoised auto-encoding networks are stacked, that is, the output of hidden layer of upper-level denoised auto-encoding network is taken as the input of the next-level denoised auto-encoding network, and Softmax classifier is added after hidden layer of the last level denoised auto-encoding network, so as to construct the stacked denoised auto-encoding network. The ferrography images of wear particle are used for unsupervised pre-training of stacked denoised auto-encoding network, network parameters are optimized by supervised fine-tuning, and ferrography images of wear particle are classified to achieve intelligent wear state recognition of mechanical equipment. The experimental results show that the stacked denoised auto-encoding network achieves the best performance when activation function is Relu, optimizer is Adam and learning rate is 0.001, and recognition accuracy is 98.43%.
- mechanical equipment /
- wear state recognition /
- ferrography image of wear particle /
- stacked denoised auto-encoding network

HTML全文

参考文献(0)

施引文献(10)

期刊类型引用(5)

1.	林俊杰. 基于状态回声网络的电梯曳引轮轮槽磨损安全性评价. 甘肃科学学报. 2023(01): 26-30+48 . 百度学术
2.	王进，王笑，韩晨霞. 基于PLC的机械设备电气安全控制方法. 机械与电子. 2022(10): 59-62 . 百度学术
3.	麻建中，王林刚，胡凯波. 基于巡检机器人的电厂设备运行状态自动识别研究. 微型电脑应用. 2022(11): 60-63 . 百度学术
4.	樊红卫，胡德顺，高烁琪，张旭辉，曹现刚. 基于深度信念网络的铁谱图像智能识别方法与试验验证. 润滑与密封. 2021(07): 15-22 . 百度学术
8.	任松，涂歆玥，朱倩雯，李眉慷. 基于深度学习的油液磨粒智能检测与分割. 振动.测试与诊断. 2024(06): 1068-1075+1243 . 百度学术

其他类型引用(5)

资源附件(0)

计量

文章访问数: 68
HTML全文浏览量: 16
PDF下载量: 15
被引次数: 10

基于堆栈降噪自编码网络的机械设备磨损状态识别

计量

出版历程

Wear state recognition of mechanical equipment based on stacked denoised auto-encoding network

期刊类型引用(5)

其他类型引用(5)

计量

出版历程

目录