[1]林永钢,吴龙跃,郑越,等.基于模型预测控制的污水系统多泵站优化调度[J].中国给水排水,2023,39(21):49-54.
　LINYong-gang,WULong-yue,ZHENGYue,et al.Optimal Scheduling Technology for Multiple Pumping Stations in Sewage Systems Based on MPC[J].China Water & Wastewater,2023,39(21):49-54.
点击复制

基于模型预测控制的污水系统多泵站优化调度

中国给水排水[ISSN:1000-4062/CN:12-1073/TU] 卷: 第39卷 期数: 2023年第21期 页码: 49-54 栏目: 出版日期: 2023-11-01
Title:
Optimal Scheduling Technology for Multiple Pumping Stations in Sewage Systems Based on MPC
作者:
林永钢1， 吴龙跃2， 郑越3， 余铭铨1， 沈大利3， 敖誉旗3， 李秀娟3， 杨彦飞1， 周永潮3
（1.中国电建集团环境工程有限公司，浙江 杭州 310000；2.中电建路桥集团有限公司，北京 100000；3.浙江大学建筑工程学院 市政工程研究所，浙江 杭州 310000）
Author(s):
LIN　Yong-gang1， WU　Long-yue2， ZHENG　Yue3， YU　Ming-quan1， SHEN　Da-li3， AO　Yu-qi3， LI　Xiu-juan3， YANG　Yan-fei1， ZHOU　Yong-chao3
（1. PowerChina Environmental Engineering Co. Ltd.， Hangzhou 310000， China； 2. PowerChina Road Bridge Group Co. Ltd.， Beijing 100000， China； 3. Municipal Engineering Research Institute， College of Civil Engineering and Architecture， Zhejiang University， Hangzhou 310000， China）
关键词:
污水泵站调度;  节能优化;  模型预测控制;  分布式模型;  多目标优化算法
Keywords:
sewage pumping station scheduling;  energy-saving optimization;  model predictive control (MPC);  distributed model;  multi-objective evolutionary algorithm
摘要:
城市污水泵站的日常优化调度对降低其能耗和管理成本具有重要的现实意义。利用模型预测控制（MPC）方法，基于分布式模型和多目标优化算法，并以污水泵站系统的能耗成本和管理成本最低为目标，提出了一种城市污水系统多泵站全局优化调度技术。采用响应面法构建预测泵站泵池液位模型，基于自适应聚类的高维多目标进化算法（ACEA）保证全局调度的速度与精度，对调度前后泵站的能耗与操作复杂度进行对比，为运维人员提供科学合理的调度策略。以W市某片区为例，针对旱天日常运行工况应用全局优化调度方法以获得可靠的调度方案。结果表明，若按优化后调度方案运行，能耗较历史方案低3%～24%，平均低约13%；复杂度要比历史方案低5%～52%，平均低约34%，说明优化方法能显著降低污水泵站运行能耗和方案的复杂性，有助于降低运行管理成本。此外，调度方案的能耗与操作复杂度间
Abstract:
The global optimal scheduling of urban sewage pumping stations is of great practical significance to reduce their energy consumption and management cost. This study utilized a model predictive control (MPC) method, based on a distributed model and a multi-objective optimization algorithm, to minimize the energy consumption and management cost of sewage pumping station systems. A global optimal scheduling technique for multiple pumping stations in urban sewage system was proposed. The response surface method was used to establish a model for predicting the pump level in pumping stations, and an adaptive clustering based evolutionary algorithm for multi-objective optimization (ACEA) was applied to guarantee the speed and accuracy of global scheduling. A comparison was made between the energy consumption and operational complexity of the pumping station before and after scheduling, which could provide scientific and reasonable scheduling strategies for operation and maintenance personnel. Taking a certain area of W city as an example, a global optimization scheduling method was applied to obtain a reliable scheduling plan for daily operating conditions in dry weather. The results showed that if the optimized scheduling scheme was operated, the energy consumption was 3%-24% lower than that of the historical schemes, about 13% on average, and the operational complexity was 5%-52% lower, about 34% on average. This indicates that the optimized method can significantly reduce the energy consumption, the complexity of the scheme and the operation and management costs. There is a certain antagonistic relationship between the energy consumption and operational complexity of the scheduling plan, indicating that decision-makers need to choose the plan reasonably based on the actual needs of energy consumption and O&M costs.
更新日期/Last Update: 2023-11-01