Scheduling charging of hybrid-electric vehicles according to supply and demand based on particle swarm optimization, imperialist competitive and teaching-learning algorithms

Increase in price of fossil fuels along with environmental issues caused by development of these resources has motivated researchers to offer resources and technologies to decrease dependency to such fuels. Hybrid-electric vehicles can receive energy from the network, store it in their battery and convert it into mechanical energy while moving. Load resulting from charging battery of these vehicles and their long charging time might increase network load and put security of the system into danger. Thus, despite various advantages, there are concerns about vast and unscheduled charging of these vehicles in the distribution network. With the increase in penetration of electric vehicles, additional load is imposed on the network due to stochastic nature of battery charging and presence of these vehicles in different places. If this load is imposed during peak load, undesirable effects occur in the distribution network including increase in losses and voltage drop. In this study, hybrid-electric vehicles connected to network and corresponding supply and demand programs are investigated; then, PSO, ICA and training-learning algorithms are used to schedule and optimize charging. Results show that in order to prevent generation of new load peak and decrease distance between peak and valley on load curve, training-learning algorithm outperforms other two algorithms. But in terms of convergence and decrease in performance cost, ICA outperforms the other two algorithms and convergence is obtained in smaller number of iterations.