Abstract
In modern generation expansion planning of power systems, installing grid‐connected renewable energy systems is preferred than thermal units due to their low generation cost and environmental pollution. However, the expanded power system must have ability to resist against any outages influenced on the frequency response of the system. So, several frequency‐constrained expansion planning models are extracted to provide a reliable infrastructure to manage the frequency behavior. The main distinction of our model with others is considering the failure of grid‐connected renewables in the expansion planning models. Furthermore, due to the lack of information about the uncertainty of malfunctions, a distributionally robust optimization approach is applied to the problem under several ambiguity radiuses. The results of implementing the proposed method on the IEEE RTS96 case show that increasing the penetration of malfunctioned units can lead to more investment on the thermal units to prevent frequency violation under any outage in the system. With increase of the Kullback–Leibler divergence from zero (stochastic) to 3 (robust), the cost of the robust model is increased about 0.02%. The model is designed for the deregulated market to increase the competition of market through maximizing their benefit and line congestion management with local marginal pricing techniques.