Optimization of Dispersed Energy Supply —Stochastic Programming with Recombining Scenario Trees

Summary

The steadily increasing share of wind energy within many power generating systems leads to strong and unpredictable fluctuations of the electricity supply and is thus a challenge with regard to power generation and transmission. We investigate the potential of energy storages to contribute to a cost optimal electricity supply by decoupling the supply and the demand. For this purpose we study a stochastic programming model of a regional power generating system consisting of thermal power units, wind energy, different energy storage systems, and the possibility for energy import. The identification of a cost optimal operation plan allows to evaluate the economical possibilities of the considered storage technologies.

On the one hand the optimization of energy storages requires the consideration of long-term planning horizons. On the other hand the highly fluctuating wind energy input requires a detailed temporal resolution. Consequently, the resulting optimization problem can, due to its dimension, not be tackled by standard solution approaches. We thus reduce the complexity by employing recombining scenario trees and apply a decomposition technique that exploits the special structure of those trees.