Data-assisted coordinated robust optimization method for urban water-power systems

With the increasing power demand of water distribution systems, utilizing pumps and water tanks as flexibility resources for the power system can effectively enhance the operational flexibility and efficiency of integrated energy systems. This paper proposes a data-assisted two-stage robust optimization method, incorporating a dynamic risk tolerance adjustment mechanism, to coordinate the operation of power and water supply systems, improve renewable energy penetration, and reduce overall operational costs. The proposed method first employs data-assisted techniques to generate uncertainty predictions and applies two-stage robust optimization to determine safe operational boundaries. Then, a convex hull method is used to construct probabilistic operation regions under different scheduling risk tolerances. Simultaneously, a dynamic risk tolerance adjustment mechanism is introduced, where the risk tolerance parameter is iteratively updated based on real-time evaluation indices to enhance the system's adaptability to uncertainties. Experimental results demonstrate that the proposed method effectively coordinates power-water system operations, enhances the flexibility of the water distribution system, improves renewable energy utilization, and reduces peak loads and reserve capacity requirements in the power system. Furthermore, the dynamic risk tolerance adjustment mechanism and probabilistic operation regions enrich scheduling strategies, enabling the system to dynamically optimize scheduling schemes based on load fluctuations and renewable energy forecasting accuracy, achieving an optimal balance between security and economy.