Optimal clustering-based operation of smart railway stations considering uncertainties of renewable energy sources and regenerative braking energies

Abstract

The smart railway stations (SRSs), as prosumer microgrids, are considered active users in smart grids. By utilizing regenerative braking energy (RBE) and renewable energy resources (RERs) along with energy storage systems (ESSs), these SRSs can participate in the prosumer market. The uncertainties of RERs in SRSs due to meteorological factors have been studied in the literature. However, there is a research gap in developing a stochastic method for optimized operating of SRSs considering the RBE uncertainties besides the RER, load, and number of passengers’ uncertainties. In this paper, a new probabilistic clustering-based framework for the optimal operation of SRSs is presented. By applying Monte Carlo Simulations (MCS), several scenarios are generated and then clustered by the k-means algorithm. The introduced method is applied to an actual SRS in Tehran Urban and Suburban Railway Operation Company. The test results of the MCS, deterministic, and proposed scenario-based approaches are compared to illustrate the proposed method. Test results imply that the related error of the scenario-based method under the real-time pricing can be less than 4.4%, while the computation time significantly decreases. Furthermore, sensitivity analysis is done to determine how the exchanging power constraints and ESS capacity might influence the SRS operation.