The operation planning of a hydroelectric sysstem, requires some random variables to be represented in a form suitable for quantitative models. Stochastic programming models require a scenario tree to be developed. In this paper a procedure to generate a scenario tree representing the time evolution of natural inflows in a hydroelectric system is presented. The procedure involves non-linear programming, linear regression techniques and deviation variables.
Keywords: Stochastic programming, scenario tree, linear regression, non-linear programming, deviation variables.
PMAPS2000: 6th International Conference on Probabilistic Methods Applied to Power Systems. Funchal, Madeira -Portugal, September 25-28, 2000.
Publicado: septiembre 2000.