Reducing the carbon footprint of urban bus fleets using multi-objective optimization

Abstract

The electrification of road vehicles was introduced as a way to significantly reduce oil dependence, increase efficiency, and reduce pollutant emissions, especially in urban areas. The goal of this paper is to find the best alternative vehicle to replace a conventional diesel bus operating in urban environments, aiming to reduce the carbon footprint and still being financially advantageous. The multi-objective nondominated sorting genetic algorithm is used to perform the vehicle optimization, covering pure electric and fuel cell hybrid possibilities (with and without plug-in capability). The used multi-objective genetic algorithm optimizes the powertrain components (type and size) and the energy management strategy. Although multiple optimal solutions were successfully achieved, a decision method is implemented to select one unique solution. A global criterion approach, a pseudo-weight vector approach, and a new multiple criteria score approach are considered to choose a preferred optimal vehicle. Real and synthetic driving cycles are used to compare the optimized buses concerning their powertrain components, efficiency and life cycle of fuel and vehicle materials. The conflict between objectives and the importance of the decision considerations in the final solutions are discussed. Passengers load and air conditioning system influence in the solutions and its life cycle is addressed.