An accelerated Benders decomposition algorithm for stochastic power system expansion planning using sample average approximation

This paper proposes a stochastic programming model and a combined solution algorithm to solve integrated resource planning (IRP) problem of electric power systems in which supply and demand side resources are combined to construct a pool of resources to expand the power systems. The problem is formulated as a two-stage recourse model, where random uncertainties in demand, operating costs, equivalent availability of generation units and customer responses to demand side management programs are taken into account. The solution methodology integrates an exterior sampling strategy, the sample average approximation algorithm, with an accelerated Benders decomposition algorithm to compute high quality solutions to the stochastic IRP problem with exponentially large number of scenarios. The proposed integrated algorithm is implemented on the modified 6, 21 and 48 bus IEEE reliability test systems and the confidence intervals of lower and upper bounds of optimal objective function as well as optimality gap are reported.