@article {pub3133pub3288,
	title = {Evolutionary Many-objective Optimization of Hybrid Electric Vehicle Control: from General Optimization to Preference Articulation},
	author = {Ran Cheng AND Tobias Rodemann AND Michael Fischer AND Markus Olhofer AND Yaochu Jin},
	year = {2017},
	month = {April},
	abstract = {A key element in hybrid car design is the energy management controller that has to guarantee peak performance for an increasing
number of conflicting objectives, e.g., fuel consumption, battery stress, emissions, noise, et al. Recently, a seven-objective controller model has been suggested to promote optimal controls of hybrid cars. Nowadays, such an optimization problem with more than three conflicting objectives is often known as a many-objective optimization problem (MaOP), which can not be solved by most multi-objective evolutionary algorithms developed for solving two- or three-objective problems, due to several challenges such as curse of dimensionality, dominance resistance and high computational cost. To meet the challenges of the seven-objective hybrid car controller optimization problem, a recently proposed evolutionary algorithm for solving MaOPs, known as the reference vector guided evolutionary algorithm (RVEA), has been applied. Our experimental results demonstrate that, with some minor modifications, RVEA is not only able to well approximate the Pareto optimal set for a many-objective hybrid car control optimization problem, but also capable of articulating user preferences in a simple and efficient manner.},
	publisher = {IEEE},
	journal = {IEEE Transactions on Emerging Topics in Computational Intelligence},
	volume = {1},
	number = {2},
	pages = {97-111}
}