[1]牛豫海,张静,张自力,等.基于VMD-LSTM的区域泵站时供水量预测[J].中国给水排水,2026,42(7):39-43.
　NIUYuhai,ZHANGJing,ZHANGZili,et al.Prediction of Hourly Water Supply in Regional Pumping Stations Based on VMD-LSTM[J].China Water & Wastewater,2026,42(7):39-43.
点击复制

基于VMD-LSTM的区域泵站时供水量预测

中国给水排水[ISSN:1000-4062/CN:12-1073/TU] 卷: 第42卷 期数: 2026年第7期 页码: 39-43 栏目: 出版日期: 2026-04-01
Title:
Prediction of Hourly Water Supply in Regional Pumping Stations Based on VMD-LSTM
作者:
牛豫海1， 张静1， 张自力1， 张娟1， 赵博洋2， 李俊江3
（1.河北建投水务投资有限公司，河北 石家庄 050022；2.辛集市建投水务有限责任公司，河北 辛集 052360；3.天津三博水科技有限公司，天津 300384）
Author(s):
NIU　Yuhai1， ZHANG　Jing1， ZHANG　Zili1， ZHANG　Juan1， ZHAO　Boyang2， LI　Junjiang3
（1. HCIG Water Investment Co. Ltd.， Shijiazhuang 050022， China； 2. Xinji Construction Investment Water Co. Ltd.， Xinji 052360， China； 3. Tianjin Sanbo Water Technology Co. Ltd.， Tianjin 300384， China）
关键词:
变分模态分解;  长短期记忆网络;  时供水量预测;  区域泵站
Keywords:
variational mode decomposition （VMD）;  long short-term memory （LSTM）;  hourly water supply prediction;  regional pumping station
摘要:
为了进一步提高泵站日常供水计划制定的科学性、合理性，提升泵站日常供水管理工作的智慧化水平，通过分析泵站时供水量数据变化规律和泵站日常调度工作的相关要求，对用水时段进行了划分，构建了长短期记忆网络（LSTM）模型，并通过变分模态分解（VMD）来提高模型性能。结果表明，VMD-LSTM模型具有理想的数据拟合和泛化能力，该模型能够实现泵站时供水量的准确预测，满足泵站日常调度管理工作的需求，可为泵站日常供水计划的制定提供数据支撑。
Abstract:
In order to further improve the scientific and rational formulation of the daily water supply plan for pumping stations and enhance the level of intelligence in their daily water supply management, this study analyzed the variation patterns of hourly water supply data from pumping stations and the relevant requirements of daily scheduling operations to divide the water usage periods. A long short-term memory (LSTM) model was constructed, and its performance was enhanced using variational mode decomposition (VMD). The results showed that the VMD-LSTM model exhibited strong data fitting and generalization capabilities. The model can accurately predict the hourly water supply of pumping stations, meet the needs of daily scheduling and management, and provide data support for the formulation of daily water supply plans.
更新日期/Last Update: 2026-04-01