Real-Time Operation of Pumping Systems for Urban Flood Mitigation: Single-Period vs. Multi-Period Optimization

Fatemeh Jafari, S. Jamshid Mousavi, Jafar Yazdi, Joong Hoon Kim

Research output: Contribution to journalArticlepeer-review

23 Citations (Scopus)

Abstract

To reduce flood risk in urban regions, it is important to optimize the performance of operational elements such as gates and pumps. This paper compares the performances of two approaches of multi-period and single-period simulation-optimization that are used to derive real-time control policies for operating urban drainage systems. The EPA storm water management model (SWMM), converting real-time rainfall data to surface runoff at network control points, i.e. pump stations, is linked to the particle swarm optimization (PSO) algorithm, evaluating the system operation performance measure (objective function) for different sets of control policies. A prototype network in a portion of the Seoul urban drainage system is used to investigate the efficiency of the proposed approaches. Results justify the high efficiency of multi-period optimization, leading to 32 and 29% average reductions in peak water level violations from a pre-defined permissible threshold at target points and the number of pump switches, respectively, in comparison with the online single-period optimization. The myopic policies derived by single-period optimization are not reliable, and in some cases, they even perform worse than ad-hoc policies applied by system operators based on their past experiences.

Original languageEnglish
Pages (from-to)4643-4660
Number of pages18
JournalWater Resources Management
Volume32
Issue number14
DOIs
Publication statusPublished - 2018 Nov 1

Keywords

  • Flood control
  • Multi-period optimization
  • Real-time operation
  • Urban drainage systems

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Water Science and Technology

Fingerprint

Dive into the research topics of 'Real-Time Operation of Pumping Systems for Urban Flood Mitigation: Single-Period vs. Multi-Period Optimization'. Together they form a unique fingerprint.

Cite this