Optimization of a Redox-Flow Battery Simulation Model Based on a Deep Reinforcement Learning Approach
Vanadium redox-flow batteries (VRFBs) have played a significant role in hybrid energy storage systems (HESSs) over the last few decades owing to their unique characteristics and advantages. Hence, the accurate estimation of the VRFB model holds significant importance in large-scale storage applicati...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-12-01
|
| Series: | Batteries |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2313-0105/10/1/8 |
| _version_ | 1827372726749233152 |
|---|---|
| author | Mariem Ben Ahmed Wiem Fekih Hassen |
| author_facet | Mariem Ben Ahmed Wiem Fekih Hassen |
| author_sort | Mariem Ben Ahmed |
| collection | DOAJ |
| description | Vanadium redox-flow batteries (VRFBs) have played a significant role in hybrid energy storage systems (HESSs) over the last few decades owing to their unique characteristics and advantages. Hence, the accurate estimation of the VRFB model holds significant importance in large-scale storage applications, as they are indispensable for incorporating the distinctive features of energy storage systems and control algorithms within embedded energy architectures. In this work, we propose a novel approach that combines model-based and data-driven techniques to predict battery state variables, i.e., the state of charge (SoC), voltage, and current. Our proposal leverages enhanced deep reinforcement learning techniques, specifically deep q-learning (DQN), by combining q-learning with neural networks to optimize the VRFB-specific parameters, ensuring a robust fit between the real and simulated data. Our proposed method outperforms the existing approach in voltage prediction. Subsequently, we enhance the proposed approach by incorporating a second deep RL algorithm—dueling DQN—which is an improvement of DQN, resulting in a 10% improvement in the results, especially in terms of voltage prediction. The proposed approach results in an accurate VFRB model that can be generalized to several types of redox-flow batteries. |
| first_indexed | 2024-03-08T11:05:13Z |
| format | Article |
| id | doaj.art-545e52dc0e9644cd8571281dd4a55374 |
| institution | Directory Open Access Journal |
| issn | 2313-0105 |
| language | English |
| last_indexed | 2024-03-08T11:05:13Z |
| publishDate | 2023-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Batteries |
| spelling | doaj.art-545e52dc0e9644cd8571281dd4a553742024-01-26T15:04:26ZengMDPI AGBatteries2313-01052023-12-01101810.3390/batteries10010008Optimization of a Redox-Flow Battery Simulation Model Based on a Deep Reinforcement Learning ApproachMariem Ben Ahmed0Wiem Fekih Hassen1Higher School of Communication of Tunis (SupCom), University of Carthage, Ariana 2083, TunisiaChair of Distributed Information Systems, University of Passau, Innstraße 41, 94032 Passau, GermanyVanadium redox-flow batteries (VRFBs) have played a significant role in hybrid energy storage systems (HESSs) over the last few decades owing to their unique characteristics and advantages. Hence, the accurate estimation of the VRFB model holds significant importance in large-scale storage applications, as they are indispensable for incorporating the distinctive features of energy storage systems and control algorithms within embedded energy architectures. In this work, we propose a novel approach that combines model-based and data-driven techniques to predict battery state variables, i.e., the state of charge (SoC), voltage, and current. Our proposal leverages enhanced deep reinforcement learning techniques, specifically deep q-learning (DQN), by combining q-learning with neural networks to optimize the VRFB-specific parameters, ensuring a robust fit between the real and simulated data. Our proposed method outperforms the existing approach in voltage prediction. Subsequently, we enhance the proposed approach by incorporating a second deep RL algorithm—dueling DQN—which is an improvement of DQN, resulting in a 10% improvement in the results, especially in terms of voltage prediction. The proposed approach results in an accurate VFRB model that can be generalized to several types of redox-flow batteries.https://www.mdpi.com/2313-0105/10/1/8energy storageredox-flow batterybattery modelingbattery state variablesparameter optimizationaccurate estimation |
| spellingShingle | Mariem Ben Ahmed Wiem Fekih Hassen Optimization of a Redox-Flow Battery Simulation Model Based on a Deep Reinforcement Learning Approach Batteries energy storage redox-flow battery battery modeling battery state variables parameter optimization accurate estimation |
| title | Optimization of a Redox-Flow Battery Simulation Model Based on a Deep Reinforcement Learning Approach |
| title_full | Optimization of a Redox-Flow Battery Simulation Model Based on a Deep Reinforcement Learning Approach |
| title_fullStr | Optimization of a Redox-Flow Battery Simulation Model Based on a Deep Reinforcement Learning Approach |
| title_full_unstemmed | Optimization of a Redox-Flow Battery Simulation Model Based on a Deep Reinforcement Learning Approach |
| title_short | Optimization of a Redox-Flow Battery Simulation Model Based on a Deep Reinforcement Learning Approach |
| title_sort | optimization of a redox flow battery simulation model based on a deep reinforcement learning approach |
| topic | energy storage redox-flow battery battery modeling battery state variables parameter optimization accurate estimation |
| url | https://www.mdpi.com/2313-0105/10/1/8 |
| work_keys_str_mv | AT mariembenahmed optimizationofaredoxflowbatterysimulationmodelbasedonadeepreinforcementlearningapproach AT wiemfekihhassen optimizationofaredoxflowbatterysimulationmodelbasedonadeepreinforcementlearningapproach |