Researchers from the University of Michigan, including Feng Guo, Guangdi Hu, Keyi Liao, Luis D. Couto, Khiem Trad, Ru Hong, Hamid Hamed, and Mohammadhosein Safari, have developed a new method for improving battery state estimation, a critical aspect of battery management systems. Their work, titled “Stability-Guaranteed Dual Kalman Filtering for Electrochemical Battery State Estimation,” was published in the IEEE Transactions on Control Systems Technology.
Accurate state estimation is essential for effective battery management, enabling optimal charging and discharging strategies, and prolonging battery life. The researchers focused on dual Kalman filtering, a technique that can simultaneously estimate both the battery’s state and parameters. However, traditional dual Kalman filtering can be prone to divergence, especially when there are large initial errors or model mismatches.
To address this issue, the team developed a Stability-Guaranteed Dual Kalman Filtering (SG-DKF) method. By using a Lyapunov-based analysis, they derived a sufficient stability condition that led to an adaptive dead-zone rule. This rule suspends parameter updates when the innovation, or the difference between the predicted and actual measurements, exceeds a certain stability bound.
The researchers applied the SG-DKF method to an electrochemical battery model and found that it achieved accuracy comparable to a dual Extended Kalman Filter (EKF). Moreover, the SG-DKF method reduced the root mean square error (RMSE) of the state of charge by over 45% under large initial state errors, demonstrating its robustness and effectiveness.
For the energy sector, particularly in applications involving battery storage and management, this research offers a promising approach to improve the accuracy and stability of battery state estimation. By enhancing battery management systems, this method could contribute to more efficient and reliable energy storage solutions, supporting the integration of renewable energy sources and improving the overall performance of energy storage systems.
This article is based on research available at arXiv.