学术报告第201430期

报告题目：Multi-layer  Model Predictive Control of Complex Regional Water Systems

报 告 人：Congcong  Sun（Polytechnic  University of Catalonia-UPC, Spain）

报告时间：11月3日（周一）上午10:30

报告地点：南一楼中311室

报告摘要:

Water,  which is an essential element for life, is becoming more and more important  because of increasing demand while decreasing supply. Water management in urban  areas is a subject of increasing concern as cities grow. According to different  control objectives and time scales, a complex regional water supply system can  be structurally organized into: Supply, Transportation and Distribution three  layers. In this work, multi-layer model predictive control (MPC) for regional  water systems is proposed. Each layer has their own MPC controller with  different time scales and control objectives. Temporal negotiation techniques  have beenused to coordinate controllers in different layers working together in  a whole part. For the non-linear relation between pressure and flow, a linear  constraint MPC based on Epanet and Constraints Satisfaction Problem (CSP) has  been applied in order to escape the high computation and non-convergence problem  appeared in non-linear controllersof large scale systems. Application has been  used to Catalonia Regional Water Network in Spain and Richmond Water Network in  United Kingdom.

报告人简介:  

Sun  Congcong received her bachelor’s degree and master’s degree from Nanjing Audit  Univerisity and Tongji University in 2008 and 2011, respectively. She is  currently a doctor’s candidate in Polytechnic University of Catalonia-UPC,  Spain, which ranks top 1 in Spain and ranks top 60 all over the world according  to the 2014 QS World University Rankings in Engineering and Technology. Her  current research interest includes the analysis and design of distributed  optimal control strategies applied to large-scale water systems management and  the efficient integrated real-time monitoring and control of drinking water  networks.