大型水利阀门轴承外圈断裂智能视觉监测方法
Intelligent Visual Monitoring Method for Outer Ring Fracture of Large Hy-draulic Valve Bearings
李飞 1周锦珂1
作者信息
- 1. 河北工程大学水利水电学院,邯郸 056038
- 折叠
摘要
水利阀门轴承外圈无法安装传感设备,其断裂缺少明确的传感阈值信号,无法通过实时传感信号完成检测.以图像识别为基础的断裂监测方法受到外圈边缘像素缺失的影响,准确性一直不高.为此,该文研究大型水利阀门轴承外圈断裂智能视觉监测方法.采集水利阀门轴承外圈图像,针对像素中的缺失值,采取低秩张量补全法填补像素.利用降噪自编码网络对阀门轴承外圈断裂像素进行特征提取.将提取到的断裂图像特征输入到支持向量机模型中,以完成外圈断裂失效状态的自动化监测.实验结果表明,该方法的特征提取效果好、状态监测准确率高.
Abstract
Sensing equipment cannot be installed on the outer ring of hydraulic valve bearing,and its fracture lacks a clear sensing threshold signal,which cannot be detected by real-time sensing signal.However,the image recognition based fracture monitoring method is affected by the missing pixels at the edge of the outer ring,and the accuracy is not high.Therefore,the intelligent visual monitoring method of fracture of bearing outer ring of large hydraulic valve is studied.The image of the outer ring of hydraulic valve bearing is collected,and the missing value in pixels is filled by low-rank tensor completion method.The feature extraction of broken pixels in valve bearing outer ring is carried out by using noise reduction self-coding network.The extracted fracture image features are input into the support vector machine model to complete the automatic monitoring of the failure state of the outer ring fracture.The experimental results show that this method has good feature extraction effect and high state monitoring accuracy.
关键词
大型水利阀门/轴承外圈断裂/状态监测/智能图像/支持向量机Key words
large water conservancy valves/bearing outer ring fracture/status monitoring/intelligent imaging/support vector machine引用本文复制引用
基金项目
河北省自然科学基金(2021)(E2021402023)
出版年
2024
NETL
NSTL
万方数据