森林工程2025，Vol.41Issue(6)：1310-1322,13.DOI:10.7525/j.issn.1006-8023.2025.06.020
林区气体压力管道泄漏超声定位自适应反卷积波束成形算法
Research on Adaptive Deconvolution Beamforming Algorithms for Ultrasonic Localization of Gas Pressure Pipeline Leaks in Forest Areas
摘要
Abstract
As an important carrier of the national new energy strategy,the leakage of gas pressure pipelines in forest areas not only causes direct economic losses,but also may lead to secondary disasters such as soil pollution,vegetation destruction and even forest fires due to the sensitivity of forest ecosystems.The current ultrasonic sound source-based lo-calization methods suffer from challenges such as high sidelobe artifact interference and insufficient localization accuracy caused by wide main lobe beamwidth in multi-leakage source scenarios.Moreover,the complex terrain,dense vegeta-tion coverage,and environmental noise in forest areas further compromise the applicability of conventional detection tech-niques.In this paper,an adaptive inverse convolution beamforming algorithm is proposed to achieve high-precision leak-age localization by optimising the weight matrix and the inverse convolution iteration strategy.Firstly,the initial weight matrix is constructed based on the minimum variance distortionless response(MVDR)criterion,and the weights are ad-justed with adaptive iteration to enhance the focusing ability of the target signal while suppressing the interference of the sidelobes.Secondly,the main lobe width is compressed through Gauss-Seidel deconvolutional iteration,thereby enhanc-ing resolution.To validate the algorithm's performance,this study establishes a pressure pipeline model with a diameter of 150 mm and operating pressure of 0.8 MPa to simulate ultrasonic signals from 0.5 mm and 0.7 mm leakage orifices,while constructing an experimental system for comparative analysis.Results demonstrate that compared with conven-tional deconvolution beamforming,the proposed algorithm reduces localization errors by 0.06 m for Source 1(0.7 mm orifice)and 0.05 m for Source 2(0.5 mm orifice)under signal-to-noise ratio(SNR)conditions ranging from-10 dB to 20 dB,while effectively eliminating artifact interference.The experiments further validate the method's robustness and computational efficiency advantages under low SNR conditions(-10 dB to 20 dB).This study provides a high-precision solution for non-contact detection of minor pressure pipeline leaks in forest environments,characterized by strong anti-in-terference capability and superior environmental adaptability.The findings hold significant implications for ensuring en-ergy transportation safety and ecological conservation.关键词
超声检测/波束成形/声源定位/气体泄漏/自适应/反卷积/阵列信号处理/麦克风阵列/声学成像Key words
Ultrasonic detection/beamforming/source localization/gas leakage/adaptive/deconvolution/array signal processing/microphone array/acoustic imaging分类
机械制造引用本文复制引用
徐敏玉,邢涛,刘鉴建县,杨阳..林区气体压力管道泄漏超声定位自适应反卷积波束成形算法[J].森林工程,2025,41(6):1310-1322,13.基金项目
黑龙江省重点研发计划指导类项目(432234015). （432234015）
