摘要
碳纤维聚氨酯(CFR-PU)夹芯板在冷藏行业中应用广泛,然而,这类产品在制作过程中容易形成聚氨酯层内气孔缺陷,直接影响其保温效果以及结构稳定性。采用敲击法实现气孔缺陷检测。由于常规K-means算法对多种振动信号的区分正确率很难超过50%,所以提出一种带约束K-means算法,以实现缺陷的定位和定量。研究搭建了一种敲击检测系统,先针对盲测中存在的4种情况进行区分试验,接着对有效检测距离进行探究,最后对缺陷大小进行定量划分。在3类试验中,本文所提出的带约束K-means方法均表现良好,缺陷识别准确率相较于常规K-means方法最高可提升6.55倍。
Carbon fiber reinforced polyurethane(CFR-PU)sandwich panels are widely used in refrigeration industry.However,during production,these products tend to form porosity defects within the polyurethane layer,which directly impact their insulation effect and structural stability.Tapping method is employed to detect porosity defects.Due to the fact that distinguishing accuracy rate of the conventional K-means algorithm on various vibration signals hardly exceeds 50%,a constrained K-means algorithm is proposed to realize localization and quantification of defects.The study establishes a tapping detection system,firstly,distinguish among four scenarios in blind tests,then,explore the effective detection distance.Finally,are the defect sizes quantitatively divided.In the three experimental categories,the proposed constrained K-means method demonstrates excellent performance,achieving a defect identification accuracy up to 6.55 times higher than that of the conventional K-means method.
作者
谭昭胜
涂君
张旭
陈涛
宋小春
TAN Zhaosheng;TU Jun;ZHANG Xu;CHEN Tao;SONG Xiaochun(School of Mechanical Engineering,Hubei University of Technology,Wuhan 430068,China;Hubei Key Laboratory of Modern Manufacturing Quality Engineering,Wuhan 430068,China)
出处
《传感器与微系统》
北大核心
2025年第9期37-40,45,共5页
Transducer and Microsystem Technologies
基金
湖北省重点研发计划项目(2022BAA075)
湖北省科技厅重大专项项目(2020ZYYD035)。
作者简介
谭昭胜(1994-),男,硕士,研究方向为无损检测新技术;通讯作者:涂君(1983-),男,博士,教授,博士研究生导师,主要研究领域为无损检测新技术。
