IElectric vehicle battery charging with safe-RL

Resumen

To become the standard power supply for electric vehicles
(EVs), Li-ion batteries need balanced current profiles in order to avoid undesirable electrochemical reactions and excessive charging times. In this work, we propose a safe exploration deep reinforcement learning (SDRL) approach in order to determine optimal charging profiles under variable operating conditions. One of the main advantages of reinforcement learning (RL) techniques is that they can learn from interaction with the real or simulated system while incorporating the nonlinearity and uncertainty derived from fluctuating environmental conditions.
However, since RL techniques have to explore undesirable states before obtaining an optimal policy, no safety guarantees are provided. The proposed approach aims at maintaining zero constraint violations throughout the learning process by incorporating a safety layer that corrects the action if a constraint is likely to be violated. Tests performed on the equivalent circuit of a li-ion battery under variability conditions show early results where SDRL is able to find safe policies while considering a trade-off between the charging speed and the battery lifespan.