Open Access Open Access  Restricted Access Subscription Access

Battery Scheduling Optimization via Reinforcement Learning under Dynamic Pricing

Sai Vinyas B S, Dr. Manish Kumar

Abstract


We address home battery scheduling to minimize electricity cost under intermittent solar generation and timevarying tariffs. A Deep Reinforcement Learning (DRL) agent (PPO) is trained on Indian grid pricing (IEX data) and solar irradiance profiles (NREL data) to decide charge/discharge actions. The RL strategy is compared against a static rule (e.g. “charge if solar available, discharge at peak price”). In simulation, the DRL agent learns to charge at low-price periods and discharge at high-price peaks, yielding 15% higher profit than the heuristic baseline. This demonstrates that PPO-based scheduling outperforms simple heuristics in dynamic pricing scenarios.


Full Text:

PDF

References


Indian Energy Exchange (IEX), “Market Data – Historical Price Data,” IEX, New Delhi, India. [Online]. Available: https://www.iexindia.com/market-data.

National Renewable Energy Laboratory (NREL), “National Solar Radiation Database (NSRDB),” Golden, CO, USA. [Online]. Available: https://nsrdb.nrel.gov/.

J. Schulman, F. Wolski, P. Dhariwal, A. Radford, and O. Klimov, “Proximal Policy Optimization Algorithms,” arXiv preprint arXiv:1707.06347, 2017.

V. Mnih et al., “Human-level control through deep reinforcement learning,” Nature, vol. 518, no. 7540, pp. 529–533, Feb. 2015.

T. P. Lillicrap et al., “Continuous control with deep reinforcement learning,” in Proc. Int. Conf. Learning Representations (ICLR), 2016.

T. Haarnoja, A. Zhou, P. Abbeel, and S. Levine, “Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor,” in Proc. Int. Conf. Machine Learning (ICML), 2018,

pp. 1861–1870.

Y. Xiong, B. Chen, J. Wang, and H. Li, “Deep Reinforcement Learning for Residential Energy Storage Scheduling,” IEEE Transactions on Smart Grid, vol. 12, no. 5, pp. 4089–4100, Sept. 2021. Available: https://www.deeplearningbook.org

S. Kell, M. K. Kouloumpis, and P. Agathokleous, “Deep Reinforcement Learning for Residential Solar-Battery Energy Management,” Applied Energy, vol. 282, 2021.

M. Tavakol-Moghaddam, A. Azadeh, and M. K. Sheikh-El-Eslami, “Deep Reinforcement Learning for Battery Energy Storage Management,” IEEE Access, vol. 8, pp. 195200–195215, 2020.

R. S. Sutton and A. G. Barto, Reinforcement Learning: An Introduction, 2nd ed. Cambridge, MA, USA: MIT Press, 2018.


Refbacks

  • There are currently no refbacks.