Enabling grid services by water distribution systems: Measurement-based modeling and optimized scheduling

Water distribution systems (WDSs) can be optimally controlled and scheduled to provide grid services, such as interruptible/curtailable load and balancing services, to the electric power system, by exploiting the flexibility of water pumps and the storage capacity of water tanks. Nonetheless, existing frameworks for modeling and optimally scheduling WDSs rely on detailed parameter knowledge, involve complex mathematical formulations, and do not account for all time horizons associated with the provision of grid services. To address these issues, this paper presents a practical two-fold methodology for enabling WDSs to provide multiple grid services. First, a new measurement-based modeling approach for WDSs is proposed to eliminate the need for parameter knowledge. The approach correlates the electric power consumption and flow rate of pumps via a convex quadratic function, enabling simpler yet more accurate modeling. Second, by leveraging this modeling approach, a comprehensive two-stage optimization framework is developed to derive the WDS operating schedule, accommodating multiple grid services across both the day-ahead and the intra-day time horizons. The proposed methodology is applied and validated on a real-world WDS in Greece. The results demonstrate the accuracy of the proposed modeling approach, the effectiveness of the proposed optimization framework, and the suitability of WDSs as flexible loads for grid services. • Simplified, measurement-based modeling approach for water distribution systems (WDSs). • WDS optimization for interruptible/curtailable load and balancing service provision. • Two-stage (day-ahead and intra-day) convex optimization, including service activation. • Application on a real-world WDS in Greece. • WDSs are suitable for grid service provision, having significant economic benefits.