| 两种用于城市用水量预测的灰色-神经网络组合模型 |
| |
| 引用本文： | 童新安,齐小刚.两种用于城市用水量预测的灰色-神经网络组合模型[J].洛阳师范学院学报,2012(2):34-38. |
| |
| 作者姓名： | 童新安 齐小刚 |
| |
| 作者单位： | 1. 西安电子科技大学理学院,陕西西安710071／洛阳理工学院数理部,河南洛阳471023
2. 西安电子科技大学理学院,陕西西安,710071 |
| |
| 摘 要： | 为了提高灰色GM(1,1)模型在城市用水量预测中的精度,结合BP神经网络的优点,给出了两种灰色-神经网络组合模型GM-BP1和GM-BP2.模型1利用神经网络对GM(1,1)模型的误差序列进行回归训练,将得到的预测值作为原始误差的修正来减小误差;而模型2由部分数据建立了GM(1,1)模型组,通过神经网络训练得到部分数据GM(1,1)模型组与真实值之间的非线性映射关系,利用这种精准的映射关系来提高预测精度.最后实际算例表明了所给方法是有效的,该组合模型可用于城市用水量的中长期预测.
|
| 关 键 词： | 城市用水量 组合模型 GM(1 1)模型 BP神经网络 误差序列 部分数据 非线性映射 |
Two Kinds of Gray-Neural Network Combined Model for the Urban Water Consumption Forecasting |
| |
| TONG Xin-an,QI Xiao-gang.Two Kinds of Gray-Neural Network Combined Model for the Urban Water Consumption Forecasting[J].Journal of Luoyang Teachers College,2012(2):34-38. |
| |
| Authors: | TONG Xin-an QI Xiao-gang |
| |
| Institution: | 1 (1.Dept.of Applied Mathematics,Xidian University,Xi’an 710071,China;2.Dept.of Math.& Phys.,Luoyang Institute of Science and Technology,Luoyang 471023,China) |
| |
| Abstract: | To improve the accuracy of GM(1,1) model used in the urban water consumption forecasting,we combine the advantages of BP neural network to propose two kinds of gray-neural network combined model GM-BP1 and GM-BP2.The main idea of GM-BP1 is as follows: firstly,error sequence using GM(1,1) model is obtained,secondly,BP neural network is utilized to train the error sequence regression,finally,the acquired prediction value is applied as the correction of original error to reduce errors;Correspondingly,in the model GM-BP2,a partial data GM(1,1) model group is gained at first,then the nonlinear mapping relationship between them and the real values is established through the BP neural network,at last the prediction is confirmed to improve forecasting accuracy.Case study validates the effectiveness of the proposed methods,and they are suitable for middle and long-term urban water consumption forecasting. |
| |
| Keywords: | urban water consumption combined model GM（1 1） model BP neural network error sequence partial data nonlinear mapping |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
