Stochastic correlated simulation: an extension of the cumulant method to include time-dependent energy sources

Evaluation of electrical power generating systems for operation and expansion planning is made traditionally with probabilistic methods, which suppose stochastic independence of the random variables of influence. For stand alone systems including time-dependent energy sources, in which power demand and generation is not dispersed, the correlation between them becomes an important factor to be considered. In this paper two ways to extend the method of cumulants in order to include correlation are proposed. The neglected chronology information is replaced using the correlation coefficient between the considered random variables. In that way, the accuracy of this method and the efficiency with respect to computation time is preserved.