To achieve greenhouse gas neutrality, the electric utilities need to integrate large amounts of intermittent renewable energy sources (RES). This integration results in high demand for energy exchange from the liberalized market according to the surpluses production or storage options. Classical generation planning assumes that the input data are deterministic, which leads to an increase in the risk potential due to the fluctuation range of this data. At the present stage, most of Generation planning techniques considering the uncertainties of input variables focus on Monte Carlo (MC) simulation and artificial neural networks (ANNs). However, MC and ANNs require comprehensive computation facilities and a big data base and also need problem-dependent modification or even integration with other techniques. These limitations make it challenging to achieve the economic operation of large-scale systems with future and spot market energy. Therefore, this paper presents integrated planning algorithm based on stochastic consideration of the uncertain input data such as the predicted consumer load, the solar radiation, wind speed, the electricity prices on the exchanges in liberalized markets depending mainly on scenario analysis.  Thereby, the optimization problem is decomposed into multi-stage decision-making process based on depicting the uncertainties in scenarios, each of which is weighted with its probability of occurrence. In this scenario analysis, the objective function consists of minimizing the annual cost over the entire scenario tree.  Due to the high demands on computing time and storage space in practical systems, decomposition approach based on Lagrange relaxation is used in this paper for solving the stochastic optimization problem. Finally, the simulation results show that the proposed stochastic optimization significantly enhances generation under high degree of uncertainties in input data.
 
Special Issue of AEIC 2024 (Electrical and System & Computer Engineering  Session)