Water Distribution System Pumping Operational Greenhouse Gas Emissions Minimization by Considering Time-Dependent Emissions Factors

Human-induced climate change caused by greenhouse gas (GHG) emissions has become a significant concern. Although water distribution systems (WDSs) provide an essential service, they also contribute to the release of GHG emissions through the use of electricity from fossil fuel sources for pumping purposes. In this paper, the reduction of both costs and GHG emissions associated with the pumping operation of WDSs is considered. Actual (time-varying) emissions factors (EFs) for the South Australia electricity grid from February 2011 to January 2012 with a 5-min time step are used to evaluate pumping operational GHG emissions and are compared with the use of an average EF, which does not consider the time dependency of EFs. An estimated (typical) 24-h EF curve, which aims to replicate the important aspects of the time dependency of actual EFs, is developed and compared for use in place of actual EFs, for when the actual variations in EFs cannot be accurately predicted for the future. Additionally, modified estimated 24-h EF curves, representing different amounts of renewable energy (wind generation) penetration, are considered to test the sensitivity of solution development to the magnitude of the variations of time-dependent EFs. Through the multiobjective optimization of pumping operations of a case study WDS, it is shown that solutions found using actual EFs can minimize GHG emissions by moving pumping to low-EF times of the day. Conversely, solutions found using an average EF can only minimize GHG emissions by pumping more consistently during the day. Additionally, solutions found using the estimated 24-h EF curve are very similar to those found using the actual EFs, suggesting that the estimated 24-h EF curve can accurately replicate the important characteristics of the time dependency of EFs and can be used in place of actual EFs to find solutions of reduced pumping operational costs and GHG emissions. Furthermore, solutions found using the modified estimated 24-h EF curves show that the development of solutions is dependent on the magnitude of the variations of time-dependent EFs.