Resilience-Oriented Multistage Scheduling for Power Grids Considering Nonanticipativity Under Tropical Cyclones

Abstract

Tropical cyclones and other weather disasters often damage transmission networks, leading to enormous threats to the safe and stable operations of power systems. To effectively confront the adverse impacts of disasters, this paper puts forward a multistage robust optimization (RO) scheduling method for regional power grids. Considering the sequential realizations of multi-uncertainties under a tropical cyclone, a multistage RO scheduling model is first established to determine the preventive strategies and the response plans before and after uncertainties, so as to realize the resilient survival of system operations from the tropical cyclone. The multistage model ensures the nonanticipativity and feasibility of the optimal robust scheduling plans. Secondly, a robust dual dynamic integer programming (RDDiP) approach is devised to handle the multistage RO problem with binary recourses. The tailored upper and lower bound optimizations approximate the cost-to-go functions with unknown analytical formulations, and the nested max-min problem is efficiently solved by iterative forward and backward passes. Computational experiments for a practical regional power grid and the IEEE 118-bus system verify the effectiveness, superiorities and scalability of the proposed RO method, and it is instructive for power system resilient scheduling against weather disasters.