eduzhai > Applied Sciences > Engineering >

Adaptive Energy Management for Real Driving Conditions via Transfer Reinforcement Learning

  • king
  • (0) Download
  • 20210505
  • Save

... pages left unread,continue reading

Document pages: 10 pages

Abstract: This article proposes a transfer reinforcement learning (RL) based adaptiveenergy managing approach for a hybrid electric vehicle (HEV) with paralleltopology. This approach is bi-level. The up-level characterizes how totransform the Q-value tables in the RL framework via driving cycletransformation (DCT). Especially, transition probability matrices (TPMs) ofpower request are computed for different cycles, and induced matrix norm (IMN)is employed as a critical criterion to identify the transformation differencesand to determine the alteration of the control strategy. The lower-leveldetermines how to set the corresponding control strategies with the transformedQ-value tables and TPMs by using model-free reinforcement learning (RL)algorithm. Numerical tests illustrate that the transferred performance can betuned by IMN value and the transfer RL controller could receive a higher fueleconomy. The comparison demonstrates that the proposed strategy exceeds theconventional RL approach in both calculation speed and control performance.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×