Optimizing Capacities of Distributed Generation and Energy Storage in a Small Autonomous Power System Considering Uncertainty in Renewables

This paper explores real power generation planning, considering distributed generation resources and energy storage in a small standalone power system. On account of the Kyoto Protocol and Copenhagen Accord, wind and photovoltaic (PV) powers are considered as clean and renewable energies. In this study, a genetic algorithm (GA) was used to determine the optimal capacities of wind-turbine-generators, PV, diesel generators and energy storage in a small standalone power system. The investment costs (installation, unit and maintenance costs) of the distributed generation resources and energy storage and the cost of fuel for the diesel generators were minimized while the reliability requirement and CO₂ emission limit were fulfilled. The renewable sources and loads were modeled by random variables because of their uncertainties. The equality and inequality constraints in the genetic algorithms were treated by cumulant effects and cumulative probability of random variables, respectively. The IEEE reliability data for an 8760 h load profile with a 150 kW peak load were used to demonstrate the applicability of the proposed method.