📄 Sciences Methods and Technologies
International Journal (SciMeTech)

Volume 2 · Issue 1 · 2026
ISSN: 3085-5284
AI-Driven Deep Learning and IoT-Based Optimization of Lithium-Ion Battery Performance with Cybersecurity Frameworks for Renewable Energy Systems
Salisu Muhmmad Lawan, Babawuro Usman, Faiza Aliyu Umar, Idris Yusif Idris, Aliyu Muhmmad Abdul, Barau Magaji
Pages 171–181 · 1. Department of Electrical/Electronic Engineering, Federal University of Technology, Babura, Jigawa State, Nigeria · 2. Department of Computer Science, Federal University of Technology, Babura, Jigawa State, Nigeria · 3. Department of Computer Engineering, Bayero University Kano, Kano State, Nigeria · 4. Equipment Maintenance Centre, Aliko Dangote University of Science and Technology, Wudi Kano State, Nigeria
⬇ PDF
Abstract
The integration of renewable energy systems requires efficient, reliable, and secure energy storage solutions. Lithium-ion batteries are widely used, but degradation, inaccurate state estimation, and cybersecurity threats limit their performance in IoT-enabled environments. This paper proposes an integrated framework combining artificial intelligence, deep learning, Internet of Things monitoring, and cybersecurity mechanisms. Battery modelling used a developed lithium-ion battery dataset generated from controlled charge–discharge experiments under varying operating conditions. The dataset was normalized and divided into 70% training, 15% validation, and 15% testing subsets, while five-fold cross-validation assessed model robustness. An LSTM model estimated state-of-charge and state-of-health, and cyber resilience was evaluated using the CICIDS2017 intrusion-detection dataset. Compared with Extended Kalman Filter and Random Forest models, the LSTM achieved 98.7% accuracy, reduced prediction error by 35%, and improved estimated battery lifespan by 23%. Complexity analysis produced O(T(hd+h²)) time and O(h²) memory, confirming suitability for real-time battery-management deployment in resource-constrained renewable-energy and smart-grid applications with manageable computational and storage requirements.
Keywords: Lithium-ion battery, Deep learning, IoT, Cybersecurity, State-of-charge, State-of-health, Renewable energy

References

  1. Li, Y., Wang, J., Zhao, H., & Chen, X. (2023). Deep learning-based state-of-charge estimation for lithium-ion batteries under dynamic conditions. IEEE Transactions on Industrial Electronics, 70(4), 4123-4134.
  2. Zhang, M., Sun, L., Liu, Q., & Wang, Y. (2023). Hybrid LSTM-Kalman filtering approach for accurate lithium-ion battery state estimation. Applied Energy, 334, 120-132.
  3. Xiong, R., Sun, F., He, H., & Pecht, M. (2022). Recent advances in lithium-ion battery modeling and state estimation. Renewable and Sustainable Energy Reviews, 158.
  4. Sharma, A., Singh, P., Gupta, R., & Verma, K. (2023). Random forest-based battery health prediction and performance analysis. Journal of Energy Storage, 55.
  5. Kim, J., Lee, S., & Park, H. (2024). LSTM-based predictive modeling for lithium-ion battery performance optimization. Energy, 275.
  6. Kumar, A., Singh, S. K., & Sharma, V. (2023). IoT-enabled smart battery management systems for renewable energy applications. IEEE Internet of Things Journal, 10(6), 5120-5132.
  7. Al-Garadi, M., Mohamed, A., Du, X., & Guizani, M. (2022). Deep learning-based intrusion detection for IoT systems: Recent advances and challenges. IEEE Access, 10, 45021-45040.
  8. Wang, H., Li, Z., & Chen, Y. (2023). Deep learning-driven battery health prognosis and remaining useful life prediction. Energy Storage Materials, 58, 210-222.
  9. Chen, X., Liu, Y., & Wang, Z. (2023). Data-driven modeling of lithium-ion battery degradation using neural networks. Journal of Power Sources, 560.
  10. Mohamed, A. E., Sallam, K. M., & El-Naggar, M. (2023). Cybersecurity challenges and solutions in smart energy systems. IEEE Access, 11, 102345-102360.
  11. He, W., Williard, N., Osterman, M., & Pecht, M. (2022). Battery prognostics and health management using artificial intelligence techniques. Journal of Power Sources, 556.
  12. Smith, K., Rahn, C., & Wang, C. (2023). Advanced control-oriented battery models for energy systems. IEEE Transactions on Energy Conversion, 38(2), 845-856.
  13. Dorri, A., Kanhere, S. S., & Jurdak, R. (2023). Blockchain and cybersecurity for IoT-enabled energy systems. IEEE Transactions on Industrial Informatics, 19(5), 4210-4221.
  14. Gupta, P., Kumar, R., & Mehta, S. (2024). AI-based anomaly detection for cyber-physical energy systems. IEEE Systems Journal, 18(1), 1021-1032.
  15. Zhang, L., Zhou, Y., & Li, H. (2023). Secure IoT frameworks for smart grid applications using deep learning. IEEE Transactions on Smart Grid, 14(3), 2105-2116.
SciMeTech · Volume 2, Issue 1 · 2026 · ISSN 3085-5284