首页|基于GA-BP神经网络的月生活需水预测——以黄河流域为例

基于GA-BP神经网络的月生活需水预测——以黄河流域为例

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
在水利高质量发展的背景下,为了更精准地反映未来需水情势,探索月尺度需水的预测方法十分必要,有利于有效规划和管理供水系统以解决水资源短缺问题.以黄河流域9个省区为例,对月生活用水规律进行分析,通过奇异谱分析(singular spectrum analysis,SSA)法对用水规律进行验证,引入遗传算法对BP神经网络预测模型进行改进,分别建立基于时间序列的单变量模型和基于影响因子的多变量模型,并将预测结果与BP神经网络模型预测结果进行比较.结果表明:改进预测模型对各省区2020年月生活需水量预测的平均相对误差基本在5%以内,精度优于BP神经网络模型,对水资源精细化管理具有一定的参考价值.
Basin monthly water demand prediction based on GA-BP neural network:a case study of the Yellow River
In the context of high-quality water conservancy development,accurate reflection of future water demand situations is of great importance and the exploration of corresponding monthly scale water demand prediction methods is necessary,which is condu-cive to effective planning and management of water supply systems to solve water resource shortages.Taking 9 provinces and regions in the Yellow River Basin as an example,the monthly domestic water use patterns were analyzed herein,and the water use patterns were verified through singular spectrum analysis.A genetic algorithm was introduced to improve the BP neural network pre-diction model.A univariate model based on time series and a multivariate model based on influencing factors were established,respectively,and the prediction results were compared with those of the BP neural network model.The results show that the aver-age relative error of the improved prediction model for predicting the monthly domestic water demand of each province in 2020 is basi-cally within 5%,indicating a better accuracy than the BP neural network model.It has certain reference value for fine management of water resources.

monthly scaledomestic waterwater demand forecastingsingular spectrum analysismachine learningBP neural net-workgenetic algorithm

沈紫菡、陈星、许钦、蔡晶

展开 >

河海大学水文水资源学院,南京 210098

水灾害防御全国重点实验室(河海大学),南京 210098

长江保护与绿色发展研究院,南京 210098

南京水利科学研究院水文水资源研究所,南京 210098

展开 >

月尺度 生活用水 需水预测 奇异谱分析 机器学习 BP神经网络 遗传算法

国家自然科学基金资助项目中央级公益性科研院所基本科研业务费专项资金资助项目中央级公益性科研院所基本科研业务费专项资金资助项目中央级公益性科研院所基本科研业务费专项资金资助项目山东省重点研发计划项目

52209031Y522001Y522018Y5200092023CXGC010905

2024

中国科技论文
教育部科技发展中心

中国科技论文

影响因子：0.466
ISSN：2095-2783
年,卷(期)：2024.19(9)
  • 11