自适应模糊神经网络系统在管道泄漏检测中的应用  

Filtering technique based on adaptive fuzzy-neural network applied to leakage detection of pipeline

作　　者：伦淑娴[1] 张化光[1] 冯健[1] 

机构地区：[1]东北大学信息科学与工程学院

出　　处：《石油学报》2004年第4期101-104,共4页Acta Petrolei Sinica

基　　金：国家自然科学基金 (No.60 2 740 1 7);沈阳市科技攻关项目 (1 0 2 30 90 2 0 0 )资助

摘　　要：输油管道泄漏信号的检测存在信噪比较低的问题 ,利用自适应模糊神经网络系统的去噪方法可以提高压力信号、流量信号的信噪比。由于自适应模糊神经网络系统具有非线性映射和自学习能力 ,能够用于噪声信号的非线性建模。它不仅能够获取信号的最佳估计 ,并且能够克服信号处理中存在的模型和噪声的不确定性、不完备性。应用结果表明 ,自适应模糊神经网络的自适应噪声抵消器不仅实现简单、节省运行时间 ,而且能快速、有效地消除流量、压力信号中的各种噪声 。There are some difficulties caused by lower signal-to-noise ratio in the signal processing of leakage detection for pipeline. It is possible to enhance the signal-to-noise ratios of pressure and flow rate signals in a pipeline by using the adaptive noise cancellation based on adaptive fuzzy neural network system (AFNNS). The AFNNS has the abilities of nonlinear mapping and self-learning property and can be used to achieve the nonlinear model of the noise. The adaptive noise cancellation based on the AFNNS can achieve an optimal reconstruction of signals and a desired robust against the effect of uncertainties in signal processing. The application of this system shows that the noise filter based on AFNNS can not only obtain simplicity of implementation and conservation of operation time, but also achieve better reconstruction performance in the pressure and flow signal processing for the leakage detection of pipeline.

关 键 词：自适应滤波 模糊神经网络系统 管道 泄漏检测 信号处理 噪声消除器 

分 类 号：TE78[石油与天然气工程]