Stochastic Multi-Timescale Power System Operations with Variable Wind Generation

Hongyu Wu,Ibrahim Krad,Anthony Florita,Bri-Mathias Hodge,Eduardo Ibanez,Jie Zhang,Erik Ela

DOI: https://doi.org/10.1109/tpwrs.2016.2635684

IF: 7.326

2017-01-01

IEEE Transactions on Power Systems

Abstract:This paper describes an integrated operational simulation tool that combines various stochastic unit commitment and economic dispatch models together that consider stochastic loads and variable generation at multiple operational timescales. The tool includes four distinct configurable sub-models within: dayahead lsecurity-constrained unit commitment (SCUC), real-time SCUC, real-time security-constrained economic dispatch (SCED), and automatic generation control (AGC). The unit commitment and dispatch sub-models within can be configured to meet multiple load and variable generation (VG) scenarios with configurable first stage and second-stage decisions determined where first-stage decisions are passed on and second-stage decisions are later determined by other sub-models in a continuous manner. The progressive hedging algorithm (PHA) is applied to solve the stochastic models to maintain the computational tractability of the proposed models. Comparative case studies, considering various configurations of stochastic and deterministic sub-models are conducted in low wind and high wind penetration scenarios to highlight the advantages of the stochastic programming during different decision-making processes. The effectiveness of the proposed method is evaluated with sensitivity tests using both economic and short-term reliability metrics to provide a broader view of its impact at different timescales and decision-making processes.