改进BP神经网络在管道泄漏检测中的应用  

Application of Improved BP Network to Leak-detecting System of Pipeline

作　　者：陈志勇[1] 李书臣[1] 翟春艳[1] 王可[1] 

机构地区：[1]辽宁石油化工大学信息与控制工程学院,辽宁抚顺113001

出　　处：《河南科技大学学报（自然科学版）》2008年第3期44-47,共4页Journal of Henan University of Science And Technology:Natural Science

基　　金：辽宁省教育厅科学研究计划资助项目(2004D031)

摘　　要：管道泄漏检测和定位在管道的安全生产中占有重要的位置。本文将用小波包分解技术提取的管道泄漏检测系统特征信号作为神经网络的输入,建立管道运行状态的神经网络分类器,根据输出对管道的运行状态进行识别。利用小波变换特性提取压力传感器的信号奇异点,根据负压力波定位法对管道泄漏点定位,仿真结果验证了该方法的有效性。The leak detection and location of the pipeline play a very important role in safe operation.This kind of characteristic signals in pipeline leak-detecting system are extracted by the wavelet analysis as the neural network's inputs in this paper,and neural network classify equipment is established for the identification of pipeline operating state through the outputs of neural network.The signal singularity of pressure sensors is extracted by wavelet-transform characteristic,and then the pipeline leaking position can be located according to negative pressure wave theory.The simulation results illustrate the effectiveness of the proposed method.

关 键 词：神经网络 输油管道 小波变换 负压力波 

分 类 号：TP389.1[自动化与计算机技术—计算机系统结构]