A Two-Stage Stochastic Unit Commitment With Mixed-Integer Recourses for Nuclear Power Plants to Accommodate Renewable Energy

To provide flexibilities for peak regulations in power systems while addressing renewable energy uncertainties, this article proposes a two-stage stochastic unit commitment model with peak regulation constraints of nuclear power plants (NPPs) under the '12-3-6-3' mode. The first stage aims to determine the optimal thermal unit commitment, and the second stage tries to adjust the operation modes of an NPP and thermal unit generation output to cope with different renewable energy scenarios. Due to the discrete adjustments of NPPs, the proposed model is mathematically a two-stage stochastic optimization with mixed-integer recourses. Furthermore, a Fenchel cut based bilevel decomposition is proposed to solve this model, where the inner second-stage subproblem is first relaxed into a linear program and a Fenchel cut is then generated to recover the facet of the convex hull solution. The technique can significantly accelerate the computational speed of the proposed model with a special mathematical structure. Numerical results on a provincial power system verify the effectiveness of the proposed model and method.