Complementary cooperative SOC/capacity estimator based on the discrete variational derivative combined with the DEKF for electric power applications

For taking the advantages of battery in the energy storage, advanced methods are required to accurately monitor and control the battery via the battery management system (BMS). This study investigates a more efficient method to increase accuracy and robustness without the high magnitude of the trans -mitted data. An alternative approach to overcome those problems, the dual extended Kalman filter (DEKF) can be selected because it can archive the good accuracy and robustness using less historical data. A major focus in DEKF is how to reflect state-of-charge (SOC) -open-circuit voltage (OCV) relation, which is the crucial characteristics of the battery. Most of the prior research has applied the SOC-OCV relation using a non-linear function such as a polynomial equation. However, since the nonlinear function is defined by the experimental data in the conventional method, the BMS needs larger storage for esti-mating the SOC and state-of-health (SOH). To overcome the limitation of the conventional DEKF, this study proposes improved DEKF combined with a discrete derivative method based on parameter iden-tification. Thus, the main objective of this study is to construct an efficient and simple SOC-OCV function using the discretization method and online parameter identification method. Experimental studies using two different types of batteries sets illustrate the high accuracy and adaptability of the proposed framework in lithium-ion battery SOC and SOH estimation. (c) 2021 Elsevier Ltd. All rights reserved.

Automated summary

This study proposes an improved DEKF method combined with a parameter identification-based discrete derivative method to construct an efficient and simple SOC-OCV function. Experimental studies demonstrate high accuracy and adaptability in lithium-ion battery SOC and SOH estimation using this approach.