Multistage Multiresolution Robust Unit Commitment With Nondeterministic Flexible Ramp Considering Load and Wind Variabilities

Abstract

This paper investigates the impacts of integrating nondeterministic flexible ramp reserves in a multistage multiresolution day-ahead robust unit commitment to cope with variability of renewable energy sources. While the common literature considers deterministic forms of flexible ramp reserves, the proposed affinely adjustable robust UC model considers explicit endogenous up and down subhourly ramp reserve deployment to effectively allocate flexible ramp capacities. In addition, unlike the conventional noncausal two-stage robust unit commitment models that immunize solutions against worst economic scenario with the dependency of decisions on the future information of uncertainty, the current nonconservative multistage model optimizes hourly commitment and dispatch scheduling for the base-case scenario, while ensuring that the specified subhourly causal ramp reserves can be adaptively adjusted. Moreover, dimensionality limitations of the conventional duality-based affinely adjustable robust optimizations are resolved. The effectiveness of the proposed model compared with conventional robust unit commitment frameworks is extensively illustrated on the IEEE 24-bus reliability test system and the IEEE 118-bus test system.