摘要
利用小波消噪原理,并通过构建RBF神经网络模型,将消噪后的信号作为RBF神经网络的输入参数,并判断该信号是否为泄漏信号。将D-S证据理论数据融合应用到天然气管道泄漏检测中,建立D-S证据理论数据融合模型。将输入的泄漏信号进行决策级综合判断,得出天然气管道泄漏的具体地点。利用C#语言开发天然气管道泄漏检测软件系统,该系统能够快速准确地识别泄漏并定位。
Making use of the wavelet denoising theory,and through establishing the RBF neural network model,the signal denoised was taken as RBF neural network's input parameter to identify the leakage signal.The D-S evidence theory data fusion was applied to natural gas leakage detection to establish its model;and the leakage signal input was judged at the decision level to locate natural gas pipeline leakage;and the C# language-based leakage detection software employed can quickly and accurately identify these leakages and can locate them.
出处
《化工自动化及仪表》
CAS
2012年第10期1268-1271,共4页
Control and Instruments in Chemical Industry
基金
国家博导联合基金资助项目--天然气管网泄漏监测技术研究(20112322110003)
关键词
小波消噪
神经网络
数据融合
证据理论
泄漏检测
wavelet denoising,neural networks,data fusion,evidence theory,leakage detection
