In this paper, we present the state-transition formulation for the unit commitment problem. This formulation is based on the definition of new decision variables, which, instead of indicating the on/off statuses of a generator, captures its state transitions between consecutive time periods. We show that this new approach produces a formulation which naturally includes valid inequalities, commonly used to strengthen other formulations. We demonstrate the performance of the state-transition formulation and observe that it leads to improved solution times especially in longer time-horizon instances. As an important consequence, the new formulation allows us to solve realistic instances of the energy industry in less than 12 minutes on an ordinary desktop PC, leading to a 45% reduction with respect to the nearest contender.
Citation
S. Atakan; G. Lulli; S. Sen, "A State Transition MIP Formulation for the Unit Commitment Problem," in IEEE Transactions on Power Systems , vol.PP, no.99, pp.1-1 doi: 10.1109/TPWRS.2017.2695964 keywords: {Mixed-integer linear programming;unit commitment}, URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7904737&isnumber=4374138