Abstract
The gradual enhancement of plug-in electric vehicle (PEV) has brought several challenges in the electricity sector in terms of infrastructures and complexities. The result of gaining popularity of PEVs is creating additional complexity on distribution system operator (DSO). However, PEVs grid to vehicle (G2V) mode charging can be considered as addition of electrical load on the system. Whereas, discharging during vehicle to grid (V2G) mode reflects PEVs as power source for the grid. Due to the PEVs uncoordinated charging, the system load increases drastically. This results in congestion scenario in the distribution network. In order to reduce the distribution system congestion, the coordinated charging strategy of G2V-V2G has been adopted in this work. To reduce the burden on the system during the coordinated operation, solar powered charging-cum-parking lot (SPCPL) has been adopted in this work. A popular machine learning approach named as random forest method (RFM) has been considered in this work in order to forecast the state-of-charge (SOC) of PEVs at the trip end. The utilization of RFM shows the proper prediction of SOC at the trip end of PEVs. Moreover, the presence of SPCPL helps the distribution system to improve the other system parameters as well. The work is tested and verified on IEEE 38 bus distribution system considering the integration of SPCPL in the industrial node using particle swarm optimization (PSO).
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have