An efficient method based on adaptive time resolution for the unit commitment problem

Unit Commitment (UC) is a computationally intensive problem, which has been solved sufficiently well for day ahead scheduling and single scenario simulations. However, the introduction of renewable energy sources and storage exposes new challenges in computation time due to the need for large-scale multi-scenario modeling while preserving inter-temporal constraints. To address this, we introduce a novel approach with adaptive time resolution that increases simulation speed and preserves accuracy. We reduce the size of the problem by grouping successive time intervals with similar net demand levels, forming a new longer interval. In comparison with the conventional UC solutions, the proposed approach is computationally more efficient, as it avoids repeated optimization for similar intervals. We analyze the quality of the solutions using the 6-bus, and IEEE 118-bus test systems as the two case-studies. The numerical results demonstrate that high quality solutions can be obtained with significant gains in computational speed, especially for the more difficult IEEE 118-bus case which is 115 times faster with a maximum error of less than ±1%.