Enhancing Lithium-ion Battery State-of-health Estimation with an IPSO-SVR Model for Sustainable Energy Practices

Strategic Insights -

Researchers at the Beijing University of Technology have developed an innovative approach to predict the health status of lithium-ion batteries, which is crucial for safe and efficient operation in applications like electric vehicles and renewable energy storage systems. By integrating multiple intelligent optimization algorithms, the IPSO-SVR model dynamically adjusts hyperparameters to capture the nonlinear aging characteristics of batteries, leading to improved prediction accuracy. Experimental results show that the proposed model outperforms a conventional PSO-SVR benchmark, achieving higher accuracy and robustness in estimating battery state-of-health.

Key Takeaways:

  • The IPSO-SVR model integrates multiple intelligent optimization algorithms (PSO, genetic algorithm, whale optimization, and simulated annealing) to enhance the accuracy and generalization of battery state-of-health estimation.
  • The model dynamically adjusts SVR hyperparameters to better capture the nonlinear aging characteristics of batteries, leading to improved prediction accuracy.
  • Experimental results show that the proposed IPSO-SVR model outperforms a conventional PSO-SVR benchmark, achieving higher prediction accuracy (mean MAE of 0.611%, mean RMSE of 0.794%, mean MSE of 0.007%, and mean R2 of 0.933).
  • The proposed approach ensures more reliable battery management, supports sustainable energy practices by enabling longer battery life spans, and promotes efficient resource utilization.
  • The IPSO-SVR model is validated using a publicly available NASA lithium-ion battery degradation dataset (cells B0005, B0006, B0007).
  • The study highlights the importance of accurate battery state-of-health estimation for safe and efficient operation of lithium-ion batteries in various applications.

Statistics:

  • Mean Mean Absolute Error (MAE) of 0.611%
  • Mean Root Mean Squared Error (RMSE) of 0.794%
  • Mean Mean Squared Error (MSE) of 0.007%
  • Mean Coefficient of Determination (R2) of 0.933

Sources:

  • Zhang, H., Shang, S., Zheng, H., Yang, F., Zhang, Y., Wang, S., Yan, Y., Cheng, J., & Xu, Y. (2025). Enhancing Lithium-ion Battery State-of-health Estimation Via an Ipso-svr Model: Advancing Accuracy, Robustness, and Sustainable Battery Management. Sustainability, 17(13), 6171.
  • VerticalNews. (2025, August 15). Investigators at Beijing University of Technology Report Findings in Sustainability Research (Enhancing Lithium-ion Battery State-of-health Estimation Via an Ipso-svr Model: Advancing Accuracy, Robustness, and Sustainable Battery Management). Ecology, Environment & Conservation, 211.