Stochastic analysis of the uncertain hourly load demand applying to unit commitment problem

In this paper, the effects of the uncertain hourly load demand are stochastically analyzed especially by the consideration of the hit rate of the Unit Commitment (UC) results. In order to minimize the effects of the actual load profile change, a new UC algorithm is proposed. The proposed algorithm calculates the UC results with the lower load level than the one generated by the conventional load forecast. In case of the worse load forecast, the deviation of the conventional UC solution can be overcome with the lower load level and the more hourly reserve requirements. This paper also presents a systematic approach for the purpose of changing commitment status for additional units to refine the UC solution with the use of the Economic Dispatch (ED) program in case of the worse load forecast applied to the UC problem. The proposed method is tested with sample systems, which shows that the new UC algorithm yields stochastically rather economical results than the conventional UC algorithms when the worse load forecast is applied. It should be noted that the hit ratio of the UC results can be enhanced by the proposed method. Therefore, it is expected that the proposed method can be used as the basic guideline for selecting the optimal load forecast applying to UC problem.