中文摘要：

提出一种新的基于无线传感器网络的管道泄漏诊断方法，以解决现有管道泄露诊断方法存在的定位精度差、距离受限和实时性差的问题。该方法采用分布式数据融合技术，在源节点处利用Symlets小波变换提取包含泄漏特征的单模态声发射信号，消除噪声干扰及频散现象；在汇聚节点处根据信号幅值大小将泄漏点两侧的信号排列组合，利用互相关时差定位和加权平均原理获得泄漏点位置坐标。实验结果表明，基于无线传感器网络的新型管道泄露诊断方法可显著提高泄漏点定位精度，其定位误差小于5％。

英文摘要：

A novel pipeline leakage diagnosis method based on wireless sensor networks is presented to improve the accuracy of leakage localization, the detecting range and the real-time performance. In order to eliminate noise and the frequency dispersion, the method uses the distributed data aggregation technique combined with the Symlets wavelet transformation to convert original signals to single-mode acoustic emission signals in source nodes, and in sink nodes, it divides the singlemode signals into couples from ordinary nodes. The leak position is obtained by the cross-correlation time differences of arrival （TDOA） localization and the weighted average method. The simulation results show that the localization accuracy can be improved obviously by aggregating the data from multi-nedes. The localization error of the method is less than 5%.