Active Balancing of Reconfigurable Batteries Using Reinforcement Learning Algorithms

Jiang, Bowen; Tang, Junfei; Liu, Yujing; Boscaglia, Luca

doi:10.1109/itec55900.2023.10187076

Cited by 3 publications

(4 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, the proposed algorithm enables the combination of classical algorithms with the proposed AQL algorithm by allowing dynamic action spaces as introduced in Equation (9). No comparison has been made with other RL algorithms, such as [18,20,22], because they are leaking in the ability to apply for different voltage levels. Therefore, per voltage level, a separate neuronal network would be required.…”

Section: Discussionmentioning

confidence: 99%

“…Furthermore, machine learning-based algorithms have shown possible use in the field of control for reconfigurable batteries. Jiang [18] proposes a RL Deep Q-Network (DQN) to control a reconfigurable battery using the three-switch topology Battery Modular Multilevel Management (BM3) [19]. They used the reconfigurable battery as a direct current (DC) source.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Controlling Algorithm of Reconfigurable Battery for State of Charge Balancing using Amortized Q-Learning

Karnehm,

Bliemetsrieder,

Pohlmann

et al. 2024

Preprint

View full text Add to dashboard Cite

Towards smart batteries for electric vehicles (EVs) smart algorithms to control battery packs, mainly reconfigurable batteries, have to be developed. This work proposes a reinforcement learning (RL) algorithm to balance the State of Charge (SoC) of reconfigurable batteries based on the topologies half-bridge and battery modular multilevel management (BM3). As RL algorithm, Amortized Q-learning (AQL) is implemented, which allows enourmous numbers of possible configurations of the reconfigurable battery to be controlled, as well as the combination of classical controlling approach and machine learning methods. This enables safety mechanisms in control. As a neural network of the AQL a Feedforward Neuronal Network (FNN) is implemented consisting of three hidden layers. The experimental evaluation using a 12-cell hybrid cascaded multilevel converter illustrates the applicability of the method to balance the SoC and maintain the balanced state during discharge. The evaluation shows a 20.3% slower balancing compared to the classical. Nevertheless, AQL shows great potential to be applied for multiobjective optimizations as an applicable RL algorithm for control in power electronics.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Controlling Algorithm of Reconfigurable Battery for State of Charge Balancing using Amortized Q-Learning

Karnehm,

Bliemetsrieder,

Pohlmann

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…It is not possible to adequately compare the proposed method to other RL algorithms, such as [18,20,22], because they cannot be applied to different voltage levels. A separate neural network would be required for each voltage level.…”

Section: Discussionmentioning

confidence: 99%

“…Furthermore, machine-learning-based algorithms have shown possible benefits in the field of controlling reconfigurable batteries. Jiang et al [18] propose an RL Deep Q-Network (DQN) to control a reconfigurable battery using a three-switch topology Battery Modular Multilevel Management (BM3) [19]. Stevenson et al [20] use a DQN to reduce the imbalance of the SoC and the current of a reconfigurable battery with four battery cells.…”

Section: Introductionmentioning

confidence: 99%

Controlling Algorithm of Reconfigurable Battery for State of Charge Balancing Using Amortized Q-Learning

Karnehm,

Bliemetsrieder,

Pohlmann

et al. 2024

Batteries

View full text Add to dashboard Cite

In the context of the electrification of the mobility sector, smart algorithms have to be developed to control battery packs. Smart and reconfigurable batteries are a promising alternative to conventional battery packs and offer new possibilities for operation and condition monitoring. This work proposes a reinforcement learning (RL) algorithm to balance the State of Charge (SoC) of reconfigurable batteries based on the topologies half-bridge and battery modular multilevel management (BM3). As an RL algorithm, Amortized Q-learning (AQL) is implemented, which enables the control of enormous numbers of possible configurations of the reconfigurable battery as well as the combination of classical controlling approaches and machine learning methods. This enhances the safety mechanisms during control. As a neural network of the AQL, a Feedforward Neuronal Network (FNN) is implemented consisting of three hidden layers. The experimental evaluation using a 12-cell hybrid cascaded multilevel converter illustrates the applicability of the method to balance the SoC and maintain the balanced state during discharge. The evaluation shows a 20.3% slower balancing process compared to a conventional approach. Nevertheless, AQL shows great potential for multiobjective optimizations and can be applied as an RL algorithm for control in power electronics.

show abstract

Design and Development of Battery Cell Protection System on LFP Battery Pack Using Bypass Techniques

Brillianto Apribowo,

Dyartanti,

Farisi

et al. 2023

2023 International Conference on Technology and Policy in Energy and Electric Power (ICT-PEP)

View full text Add to dashboard Cite

Active Balancing of Reconfigurable Batteries Using Reinforcement Learning Algorithms

Cited by 3 publications

References 19 publications

Controlling Algorithm of Reconfigurable Battery for State of Charge Balancing using Amortized Q-Learning

Controlling Algorithm of Reconfigurable Battery for State of Charge Balancing using Amortized Q-Learning

Controlling Algorithm of Reconfigurable Battery for State of Charge Balancing Using Amortized Q-Learning

Design and Development of Battery Cell Protection System on LFP Battery Pack Using Bypass Techniques

Contact Info

Product

Resources

About