Search

In this paper, security constrained generation scheduling (SCGS) problem for a grid incorporating thermal, wind and photovoltaic (PV) units is formulated. The formulation takes into account the stochastic nature of both wind and PV power output and imbalance charges due to mismatch between the actual and scheduled wind and PV power outputs. A hybrid technique in which the basic elements are a genetic algorithm (GA) with artificial neural network (ANN) and a priority list (PL) is used to minimize the total operating costs while satisfying all operational constraints considering both conventional and renewable energy generators. Numerical results are reported and discussed based on the simulation performed on the IEEE 24-bus reliability test system. The results demonstrate the efficiency of the proposed approach to reduce the total production cost for real time operation. Moreover, the results verified that the proposed approach can be applied to different problem dimensions and can score more favorably compared with analytical techniques. [ABSTRACT FROM AUTHOR]

Highlights: [•] Increasing renewables reduce “residual demand” to be met by conventional power plants. [•] MILP faces computational difficulties for low load UC problems. [•] New EPL method is developed specifically for low load UC. [•] The EPL method proves to be fast, accurate and close to optimal. [ABSTRACT FROM AUTHOR]