Multi-timescale risk-averse restoration for interdependent water–power networks with joint reconfiguration and diverse uncertainties

Abstract

The growing interdependencies between water and power systems have increased the risk of cascading disruptions and widespread blackouts. Such interdependencies, together with different operational characteristics and multiple uncertainties, introduce additional complexities to service restoration. To address these issues, this paper proposes a coordinated multi-timescale restoration strategy for interdependent water–power networks (IWPNs). First, we model the IWPN as network-based with physical mechanisms, incorporating component-wise interdependencies and varying consumer demands. Features comprising pipe damage (water leakage) and storage as well as renewable generations are modelled to better reflect restoration better. Specifically, the joint reconfigurability of water and power networks is first applied for adjustment of topologies and leverages off backup components by coordinated setting of valves and switches. Then, an updated estimation for multiple uncertainties during restoration is utilized, which offers increasing clarity to support better decision-making. These uncertainties arise from renewable generations and water and power demands. A multi-timescale decision framework is developed to capture these effects and tune restoration measures based on response speeds to facilitate efficient and reliable restoration. Finally, the method is implemented by combining robust optimization and risk-averse stochastic programming and applied to a community-scale test system with 25 water nodes and 33 power buses. The proposed method is compared with five conventional methods with numerical results demonstrating the improvements arising from an interdependent restoration, joint reconfigurability, and multi-timescale optimizations.