Electric vehicle range prediction for constant speed trip using multi-objective optimization

• Published in: Journal of power sources Vol. 275; pp. 435 - 446
• Main Authors: Vaz, Warren, Nandi, Arup K.R., Landers, Robert G., Koylu, Umit O.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.02.2015
• Subjects: Constants; Driving strategy; Electric batteries; Electric charge; Electric potential; Electric vehicle; Electric vehicles; Genetic algorithm; Multi-objective optimization; Optimization; Range prediction; Strategy; Voltage; Driving strategy; Electric vehicle; Multi-objective optimization; Genetic algorithm; Range prediction
• ISSN: 0378-7753; 1873-2755
• DOI: 10.1016/j.jpowsour.2014.11.043

Due to the limited range and long charging time for electric vehicles, proper utilization of the stored battery energy is crucial. Current methods for electric vehicle range estimation do not help the driver to formulate a driving strategy based on trip parameters (e.g., trip speed) related to power savings. This can be done by predicting the driving range based on optimal trip parameters prior to the trip enabling the driver to formulate a suitable driving strategy. This study proposes a novel strategy that presents a number of optimal trip speeds to the driver, along with the total trip time corresponding to a predicted range. The optimal speeds were obtained by solving a multi-objective optimization problem that maximized electric motor efficiency and minimized power consumption. Two approaches to calculate the objective functions were considered: using constant battery voltage and using battery voltage as a function of the state-of-charge. Pareto-optimal fronts were obtained and a plot of the predicted range and trip times for optimal speeds was created. It was found that the shape of the fronts was not affected by the approach; however, the range was overestimated when a constant battery voltage was used. •A multi-objective approach to range prediction of electric vehicles is proposed.•Electric motor efficiency and power consumption were considered as objectives.•Pareto-optimal fronts were obtained using NSGA-II; “knee” used for decision-making.•Range and trip time plot for different initial battery SOC values.•Proposed strategy is easily implemented, offers flexibility, helps in trip planning.