Accurate estimation of the battery State of Charge (SOC) is essential for efficient energy management in battery-operated systems. Recursive Bayesian estimation techniques are commonly employed for this purpose due to their resilience in managing uncertainty and their ability to adjust estimates with fresh measurements. Particle filters, in particular, are well-suited for non-linear dynamic systems such as lithium-ion batteries because of their flexibility, which allows for more accurate and reliable estimates compared to typical Kalman filter-based methods that assume Gaussian noise. In this study, we utilize a particle filter with systematic resampling, resulting in increased accuracy in SOC estimates. Our proposed approach demonstrates superior performance over the particle filter with stratified resampling, exhibiting reduced Root Mean Square Error (RMSE) across various conditions including dynamic load changes and temperature fluctuations.
Particle Filter with Systematic Resampling for Battery State of Charge Estimation
31.07.2024
1030983 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Resampling Based Particle Filter Estimation of a Quadrotor
Springer Verlag | 2023
|Interacting multiple model particle filter optimization resampling algorithm
British Library Online Contents | 2017
|