论文部分内容阅读
利用对角递归网络能够描述非线性动态系统的特点,设计了用于气体管道泄漏检测定位的动态网络模型。网络的训练样本由管道各种工况下的压力、流量数据组成。试验结果表明,与传统的静态前馈神经网络模型相比,本研究建立的网络能更好的反映气体在管道中的流动特性,实现气体管道的泄漏检测与定位。
Using the diagonal recursive network to describe the characteristics of nonlinear dynamic system, a dynamic network model for gas pipeline leak detection and localization is designed. The training samples of the network consist of pressure and flow data under various working conditions of the pipeline. The experimental results show that compared with the traditional static feedforward neural network model, the network established in this study can better reflect the flow characteristics of gas in the pipeline and detect and locate the gas pipeline leakage.