Modelling an energy demand management system using stochastic optimization method

In this thesis, an energy and reserve scheduling model has been designed using stochastic optimization method. The model has been designed for a microgrid structure where renewable energy generation is used to feed the electricity demand of customers. Demand side response from the electricity customers is used for energy and reserve planning. Model is formulated as a two-stage stochastic linear programming problem and is solved using the L-Shaped Method. In addition, another optimization model is designed in order to schedule the household loads of residential demand side participants. This model is formulated as a linear integer programming problem and is solved using the Branch and Bound algorithm. Uncertainties related to the operation of the microgrid has been modelled using scenarios in the stochastic programming framework. Uncertainties that are considered in the model are; solar energy generation, wind energy generation, energy demand levels and the cost of unbalancing. In this regard, wind speed, solar irradiation and energy demand data has been analyzed and related probability distribution parameters have been calculated. Monte Carlo sampling is used to generate random values on several intervals. These values have been used to formulate the stochastic scenarios of the model. Other scheduling methods have been used for comparison in order to measure the efficiency of the stochastic scheduling model. Scheduling results for a deterministic planning method and a no reserve planning approach has been considered for the comparison. Scheduling results of all models have been subjected to different realization scenarios. Analysis of the results has shown that the stochastic scheduling method yields an economic advantage compared to other methods.