A cloud-supported plug-in hybrid electric buses energy-saving control integrating spatial–temporal traffic condition

The lack of adaptive control parameters combined with fluctuating traffic conditions complicates energy-saving optimization for plug-in hybrid electric buses (PHEB). This paper proposes an innovative updatable and evolvable energy-saving control based on vehicle-to-cloud (V2C) and traffic condition. We construct the spatial–temporal characteristics by extracting the representative route cycles in different traffic periods. Then, we establish an updatable and evolvable energy-saving control strategy integrating spatial–temporal traffic conditions, including an online updating strategy and an evolving torque distribution strategy. According to the spatial–temporal feature trigger mechanism, some key parameters, including the driving mode parameter and double deep Q learning (DDQL) network parameter, are dynamically updated to the vehicle controller by V2C. We verify the strategy on a vehicle-cloud simulation platform, and results show that the strategy enhances energy efficiency and adaptability compared with other conventional methods, improved by 14.13%, 11.72%, and 10.18% in the morning peak, off-peak, and evening peak.