Real-time charging coordination of plug-in electric vehicles based on hybrid fuzzy discrete particle swarm optimization

Abstract

The main impact of uncoordinated plug-in electric vehicle (PEV) charging is adding new time-variant loads that can increase the strains on the generation units, transmission and distribution systems that may result in unacceptable voltage drops and poor power quality. This paper proposes two dynamic online approaches for coordination of PEV charging based on fuzzy genetic algorithm (FGA) and fuzzy discrete particle swarm optimization (FDPSO). The algorithms will minimize the costs associated with energy generation and grid losses while also maximizing the delivered power to PEVs considering distribution transformer loading, voltage regulation limits, initial and final battery state of charges (SOCs) based on consumers' preferences. The second algorithm relies on the quality and speed of DPSO solution for more accurate and faster online coordination of PEVs while also exploiting fuzzy reasoning for shifting charging demands to off-peak hours for a further reduction in overall cost and transformer loading. Simulation results for uncoordinated, DPSO, FGA and FDPSO coordinated charging are presented and compared for a 449-node network populated with PEVs. Results are also compared with the previously published PEV coordinated charging based on maximum sensitivity selections (MSS). Main contributions are formulating the PEVs charging coordination problem and applying different optimization methods including online FGA and FDPSO considering different driving patterns, battery sizes and charging rates, as well as initial SOCs and requested final SOCs.