作　　者： (王兆吉）;

机构地区： 河北省保定水文水资源勘测局,河北保定071000

出　　处： 《水科学与工程技术》 2017年第4期35-38,共4页

摘　　要： 城市需水量预测是水资源可持续发展的研究基础。需水量预测考虑的影响因素较复杂,增加了需水量预测难度。通过建立RBP神经网络模型,以河北省A城市为例,进行城市需水量拟合与预测,与传统BP神经网络模型和灰色系统模型计算结果进行对比分析,结果表明RBP神经网络模型拟合的相对误差为2.65%,模型预测结果的相对误差为3.92%,计算结果精度高于另外两种方法,对今后城市需水量预测方法研究提供了一种有效方法的借鉴。 The prediction of urban water demand is the basis of sustainable development of water resources. The influence factors of water demand prediction are complex, and the difficulty of water demand forecasting is increased. Through the RBP neural network model is established in A city of Hebei Province as an example, the city water demand of fitting and prediction results were compared with the traditional BP neural network model and grey system model. The result shows that the relative error of RBP neural network model is 2.65%, relative error of model predictions was 3.92%, calculation results precision is higher than the other two methods, it provides an effective method for future reference to forecast method of city water.

关 键 词： 需水量预测 神经网络 神经网络 灰色系统模型 方法比较