Energy consumption and battery aging minimization using a Q-learning strategy for a battery/ultracapacitor electric vehicle

Propulsion system electrification revolution has been undergoing in the automotive industry. The electrified propulsion improves energy efficiency and reduces dependence on fossil fuel. However, batteries of electric vehicles experience degradation process during vehicle operation. Research considering both battery consumption battery/ultracapacitor is still lacking. This study proposes a Q-learning-based strategy to minimize consumption. Besides Q-learning, two rule-based management methods are also proposed optimized using Particle Swarm Optimization algorithm. A model first presented, where severity factor considered experimentally validated with help Genetic Algorithm. In results analysis, Q-learning explained optimal policy map after learning. Then, result from without ultracapacitor used as baseline, which compared strategies. At learning validation driving cycles, indicate that slows down by 13–20% increases range 1.5–2% baseline ultracapacitor.