ELECTRIC CAPACITY EXPANSION UNDER UNCERTAIN DEMAND - DECOMPOSITION APPROACHES

We present a stochastic modeling for electric capacity expansion plann ing under uncertainty in demand. The goal of this problem is to determ ine the most interesting investments (plants and capacity levels) over the considered planning time (up to several years). Periods are divid ed into smaller subperiods (e.g. weekly or monthly) for which demand i s assumed uncertain and modeled as a continuous probability distributi on function. This leads to consider the risk associated to each decisi on for the capacity to be used (electricity generation). A first appro ach as a non-linear continuous model is presented. Benders Decompositi on, and Lagrangean Relaxation-Decomposition are proposed as solving me thods, where the structures of the related subproblems are exploited t o speed up the convergence. We provide a large computational experienc e and comparisons within these methods and other general purpose optim ization packages, and focus the report in the advantages of each.