Year: 2025
Type: article
Abstract: Water distribution systems (WDSs) are increasingly required to respond to dynamic financial and regulatory signals. This study presents a computationally tractable multi‐objective optimization framework for minimizing time‐variant electricity costs, carbon intensity, and water age. We apply this framework in a large, complex WDS over monthly time periods to demonstrate computational tractability and describe tradeoffs in energy costs, carbon emissions, and water quality under realistic operating and billing conditions. We achieve computational tractability by combining a novel model reduction approach with search space reduction (domain targeting) and algorithmic efficiency tools (search‐integrated feasibility prescreening). The result was a 43% reduction in hydraulic simulation time and a 20% reduction in water quality simulation time for our case study system. By modifying a combination of tank level and time control set points in our case study, we identify opportunities for reducing energy costs by up to 5% without compromising GHG emissions and water age and up to 8% with increases in water age. This work underscores the importance of multi‐objective formulations for dynamic WDS optimization and the imperative of computationally efficient optimization workflows for practical application in large systems with monthly billing cycles.
Source: Water Resources Research
DOI: 10.1029/2024wr039830
Authors: E. Musabandesu, M. S. Mauter
Institutions: Stanford University, SLAC National Accelerator Laboratory
Language: English
Subfield: Civil and Structural Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $3,000
Awards: G3150238734, G4986763108
OpenAlex ID: W4413995491