Design of Water Distribution Systems with Behind-the-Meter Solar under Uncertainty: Joint Impact of Optimization Approaches and Robustness Metrics

Water distribution systems (WDSs) are critical water infrastructure for a reliable water supply. To reduce dependence on centralized electricity grids and lower greenhouse gas emissions, behind-the-meter (BTM) solar photovoltaic systems are increasingly installed as supplementary energy sources for WDSs. A robust WDS design integrating BTM solar systems that can perform consistently under uncertain future conditions is essential to maintain future water supply reliability. There are different approaches to achieving a robust design, including postoptimization robustness evaluation (PORE) and a robust optimization (RO) approach, both of which rely on the use of a robustness metric to evaluate the consistency in the performance of solutions under future uncertainty. This study investigates the joint impact of the optimization approach and robustness metric used on the robust design outcomes via two case studies. Results show that the computationally intensive RO approach typically yields more “conservative” designs, ensuring superior energy and hydraulic performance with marginal increases in lifecycle costs. In addition, the use of robustness metrics with overly high levels of risk aversion, such as those emphasizing the deviations in system performance, can significantly constrain the feasible solution space for both approaches.