نوع مقاله : مقاله پژوهشی
عنوان مقاله English
نویسندگان English
The growing adoption of electric vehicles (EVs) in urban transportation networks, combined with battery capacity limitations, the dependence of driving range on the state of charge (SOC), and the need for access to charging infrastructure, has rendered conventional routing and traffic assignment methods insufficient to fully represent the actual behavior of these vehicles. This study aims to develop an integrated framework for determining shortest paths and assigning traffic flows for electric vehicles while simultaneously accounting for energy constraints, battery state of charge, and the sequence of activities within a trip chain. The trip chain is considered the unit of decision-making, and feasible routes must satisfy not only travel cost and time criteria but also energy feasibility throughout all stages of the trip. Modeling and computational implementation were carried out in MATLAB. A multi-label shortest-path algorithm incorporating energy constraints and activity sequencing was developed to identify energy-feasible routes, while network-level flow assignment was performed using the user equilibrium framework. A genetic algorithm was employed to search the solution space and identify an appropriate combination of routing and energy-related decisions, and a modified Frank–Wolfe framework was used to achieve network flow equilibrium. The model's performance was evaluated using the Nguyen and Wang test networks under various scenarios, including the base case, increased charging capacity, reduced driving range, changes in trip chains, and smart charging. The results showed that the smart charging scenario reduced the average total travel time from 42.6 to 36.8 minutes (a 13.6% improvement). The generalized time–energy cost decreased from 100 to 86 units. Total network energy consumption declined from 11.9 to 9.5 MWh (a 20.3% reduction), while the trip-chain failure rate decreased from 14.8% to 5.9%.
کلیدواژهها English