基于Elman神经网络城市供水管网漏损预测模型研究

Study on Forecasting Models for Urban Water Supply Network Based on Elman Neural Network

管道漏子的发生情况是进行漏损控制的基础,为了对管道漏子数进行有效预测,提出了一种基于Elman神经网络预测方法,采用自适应学习速率动量梯度下降反向传播算法,显著提高了网络的训练速率,有效地抑制了网络陷入局部极小点。文中分别采用Elman神经网络、BP神经网络和RBF神经网络对某城市外环线DN300管道漏子发生数时间序列进行仿真预测,经比较,证明前者具有收敛速度快、预测精度高的特点,表明利用Elman回归神经网络建模对管道漏子数进行预测是可行的,能为管网维护,管道更新提供有效依据。

pipe leakage forecasting is the basis of control method on pipe leakage.In order to improve the precision of forecasting,an Elman artificial neural network(ANN) approach for forecasting is proposed.In the training algorithm of the network,a back-propagation algorithm with adaptive learning speed and momentum gradient falling is used,which can obviously improve the training speed of the network and effectively prevent the network to trap in local minimum.The forecasting model tested by actual time series data on pipe leakage of DN300 from some a city is established by using Elman neural network,BP neural network and RBF neural network.By analyzing and comparing,the former features quick convergence speed and high forecasting precision.The simulation results show that the method is feasible,which can effectively improve the precision of pipe leakage forecasting and provide the basis to network maintenance and pipe renewal.