Generation and Transmission Expansion Planning Towards a 100% Renewable Future

This paper proposes a novelmodeling framework and decomposition-based solution strategy combining stochastic programming (SP) and robust optimization (RO) to dealwith multiplex uncertainties in coordinated mid- and long-term power system planning. The problem is formulated as a multi-year generation and transmission planning problem from an independent system operator (ISO)'s perspective to minimize both expansion and operational costs under binary and continuous uncertainties, i.e., system component contingency and load/generation variation. N-k contingencies are captured in RO using the reformulated contingency criteria, while the load/generation uncertainty is considered in SP embedded with RO using operating scenarios generated from the historical data with spatiotemporal correlations. The original hybrid model is highly intractable, but the intractability can be relieved by the proposed decomposition strategy based on the column-and-constraint generation and L-shaped algorithms. We apply our model to perform long-term system planning under extremely high renewable penetration and investigate the case of 100% renewables in long-term planning. Numerical experiments on multi-scale test systems verify the efficacy of the proposed approach.

Automated summary

This paper proposes a novel modeling framework and decomposition-based solution strategy to address multiplex uncertainties in coordinated mid- and long-term power system planning using stochastic programming and robust optimization. The proposed approach considers both system component contingency and load/generation variation to minimize expansion and operational costs.