Short-term thermal generation scheduling with application of probabilistic and deterministic methodologies

The authors present a method of calculating short-term thermal plant generation scheduling using both probabilistic and deterministic methods, by using each method in sequence. First, the probabilistic approach, which takes into account uncertainties on both generation and demand sides, together with dynamic programming is used to determine the scheduled thermal units. Then, the deterministic approach is applied, through a dispatching technique, to determine the generation and reserve contribution from each generating unit. A multi-state Markov model is used to cope with the uncertainties on the generation side. A normal distribution function specified by the mean and standard deviation values, is employed to represent the demand uncertainties. With the proposed method, the optimum schedules can be obtained without losing any stochastic property considerations on both the generation and demand sides. The simulation is carried out with a small fictitious power system and the results show the benefits of the proposed method compared to a conventional method using only a deterministic approach