A particle swarm optimization based control strategy for plug-in hybrid electric vechicles at residential networks level

With the increasing awareness on environmental protection and energy resources reservation, plug-in hybrid electric vehicles (PHEVs) are being developed and deployed around the world. However, the large-scale deployment of PHEVs will impose a great burden on the residential power grid, so effective control strategies, especially those for vehicle-to-grid (V2G), are much needed. In this paper, the stochastic behaviors of PHEVs are examined in real scenarios, and the V2G process is modeled considering various limits. Possible exchange power trajectories are enumerated and pruned based on the proposed load distance for each PHEV in order to generate the optimized solution space. Particle swarm optimization (PSO) is applied for several time-steps with the updating of load profile and PHEV groups for global optimization. Some simulation studies are carried out with different fitness functions considering the load variance, cost and comfort of PHEV users. Simulation results show that the proposed optimization method is able to shave the residential peak load effectively, and its performance with different fitness functions are analyzed and compared.