Page 249 - 《应用声学》2025年第1期

第 44 卷 第 1 期 Vol. 44, No. 1
2025 年 1 月 Journal of Applied Acoustics January, 2025

⋄ 研究论文 ⋄

基于LabVIEW的智能建筑外墙饰面砖脱空

识别软件开发 ∗

龙士国 † 刘昳嵘 李心怡 欧阳德威 彭 强 彭克军

(湘潭大学土木工程学院 湘潭 411105)
摘要：建筑外墙饰面砖出现脱空现象导致行人出行安全隐患。建筑外墙饰面砖常用锤击检测法识别脱空，识
别结果需检测人员的判断。为了进一步提高检测准确率，该文应用 LabVIEW 开发了智能建筑外墙饰面砖脱
空识别软件。该软件基于有限状态机模型开发了数据采集、数据分析和数据管理等功能模块。其中数据分析
模块通过提取首波幅值和小波系数积分比特征，运用拉依达准则对数据进行预处理，结合 K 近邻算法构建脱
空识别模型来判别饰面砖是否脱空。并通过制作外墙试件进行实验验证，结果表明软件识别试件的脱空识别
率达到 92%。因此，该软件能有效识别建筑外墙饰面砖脱空的大小和位置，且运行稳定可靠，在实际工程中有
较高的应用价值。
关键词：建筑外墙饰面砖；脱空；LabVIEW；软件
中图法分类号: TP311.1 文献标识码: A 文章编号: 1000-310X(2025)01-0245-07
DOI: 10.11684/j.issn.1000-310X.2025.01.026

Development of intelligent exterior tile wall debonding identification software

based on LabVIEW

LONG Shiguo, LIU Yirong, LI Xinyi, OUYANG Dewei, PENG Qiang and PENG Kejun

(School of Civil Engineering, Xiangtan University, Xiangtan 411105, China)

Abstract: Debonding of exterior wall tiles can present safety hazards to pedestrians. Although the conventional
method of identifying tile debonding is through hammer tapping, it still requires expert judgement, leading to
inefficiency. This paper introduces a smart software developed using LabVIEW to accurately detect debonding
in exterior wall tiles, improving detection precision. The software operates on a finite state machine model that
includes functional modules for data acquisition, analysis, and management. The data analysis module extracts
features such as the first wave amplitude and integral ratio of wavelet coefficient. The data is preprocessed
with the Pauta criterion and combined with the K-nearest neighbors algorithm to formulate a debonding
detection model that determines debonding defects in the exterior wall tiles. Experimental verification using
fabricated exterior wall specimens showed that the software recognizes 92% of the deflated recognition rate of the
specimens. In conclusion, the software accurately identifies the size and location of debonding in exterior wall
tiles, providing consistent and reliable performance, and significant practical value in engineering applications.
Keywords: Exterior wall tiles; Debonding; LabVIEW; Software

2023-09-16 收稿; 2023-12-27 定稿
国家自然科学基金项目 (51908481)
∗
作者简介: 龙士国 (1972– ), 男, 湖南湘潭人, 博士, 教授, 研究方向: 工程安全评估及智能控测。
† 通信作者 E-mail: longsg@xtu.edu.cn