Fast stochastic security-constrained unit commitment using point estimation method

Security-constrained unit commitment (SCUC) is a key component of operating electricity markets. By increasing the share of renewable energies on the generation side, and emergence and growth of new stochastic loads on the demand side, stochastic SCUC has become more important for secure-optimal operation of the market. Scenario-based techniques have been suggested widely for stochastic SCUC in the literature. However, they are usually very time-consuming. This problem is escalated in large-scale power systems with high penetration of stochastic generation and loads. To mitigate the computational burden of SCUC problem, this paper develops an algorithm, based on point estimation method and Bender's decomposition technique. The proposed approach breaks the probabilistic problem into a few deterministic points with much lower computation burden yet with minimal loss of accuracy. The proposed approach is implemented on a six-bus system as the first numerical study, and on a modified IEEE 118-bus system with 94 probabilistic variables as the second case study. The efficacy of proposed algorithm is confirmed, especially in the last test case with notable reduction in computational burden without considerable loss of precision. Copyright (c) 2015 John Wiley & Sons, Ltd.