Page 162 - 《应用声学》2022年第1期
P. 162
第 41 卷 第 1 期 Vol. 41, No. 1
2022 年 1 月 Journal of Applied Acoustics January, 2022
⋄ 研究报告 ⋄
声发射衰减特性管道泄漏定位方法 ∗
张 曦 章兰珠 †
(华东理工大学机械与动力工程学院 上海 200237)
摘要:声发射技术具有灵敏度高、实时性强、覆盖范围大等优点。泄漏产生的声发射波沿管壁传播会发生衰减,
通过研究衰减系数与金属晶粒散射和热流损失的关系,建立准确的声发射能量衰减模型。在此基础上,针对声
发射频带宽的特点对传统衰减定位模型进行改进,提出宽频带声发射源定位模型,该方法先通过实验确定泄
漏信号频带,再将滤波后的信号经过小波包分解,提取小波包能量特征进行定位。针对无法完全滤除的噪声
提出三传感器定位模型,利用 3 组信号互相做差可减小噪声的影响。实验表明:采用宽频带定位模型比传统定
位模型更具优势,定位误差平均减小 5%,低泄漏率 (60 mL/min) 时可减小 8%;增加一个传感器后定位误差为
5%,几乎不受泄漏率影响,证明了该文定位模型的准确性。
关键词:管道泄漏;声发射;衰减模型;小波包分解;定位模型
中图法分类号: TB559 文献标识码: A 文章编号: 1000-310X(2022)01-0158-10
DOI: 10.11684/j.issn.1000-310X.2022.01.019
The pipeline leakage location method based on acoustic emission
attenuation characteristics
ZHANG Xi ZHANG Lanzhu
(School of Mechanical and Power Engineering, East China University of Science and Technology, Shanghai 200237, China)
Abstract: Acoustic emission (AE) technology has the advantages of high sensitivity, strong real-time and large
coverage. The acoustic emission wave generated by leakage will be attenuated along the pipe wall. By studying
the relationship between attenuation coefficient and metal grain scattering and heat flow loss, an accurate
acoustic emission energy attenuation model is established. On this basis, according to the characteristics of
acoustic emission radio frequency (RF) bandwidth, the traditional attenuation localization model is improved,
and a broadband acoustic emission source localization model is proposed. Firstly, the frequency band of leakage
signal is determined by experiment, and then the filtered signal is decomposed by wavelet packet to extract
the energy feature of wavelet packet for location. Aiming at the noise that can not be completely filtered out,
a three-sensor positioning model is proposed. The influence of noise can be reduced by using three groups of
signals to make difference with each other. The experimental results show that: the broadband positioning
model has more advantages than the traditional one, the positioning error is reduced by 5% on average, and it
can be reduced by 8% when the leakage rate is low (60 mL/min); after adding a sensor, the positioning error
is 5%, and it is almost not affected by the leakage rate, which demonstrates the accuracy of the positioning
model in this paper.
Keywords: Pipeline leakage; Acoustic emission; Attenuation model; Wavelet packet decomposition; Localiza-
tion model
2021-01-11 收稿; 2021-04-07 定稿
国家重点研发项目 (2018YFC0808600)
∗
作者简介: 张曦 (1995– ), 女, 重庆人, 硕士研究生, 研究方向: 信号分析。
† 通信作者 E-mail: lzzhang@ecust.edu.cn