Autonomous Characterization of Lithium-Ion Battery Model Parameters utilizing a Mathematical Optimization Methodology

Kalman filtering is commonly used for state-of-charge (SOC) estimation for lithium-ion (Li-ion) batteries owing to its simplicity, computational efficiency, and relatively precise results. However, kalman filters depend on the Li-ion battery model. Several laboratory tests such as incremental current and dynamic stress tests are required to determine battery model parameters in model-based SOC estimation. These tests such as incremental current test and dynamic stress test are time-consuming and can take multiple days. A mathematical optimization along with a battery test method, which does not need rest time for battery, are adopted to reduce the battery parameter identification time, drastically. A mathematical optimization stage is embedded prior to Kalman Filter based SOC estimation computing the battery open circuit voltage (OCV) and as well as an initial guess of the RC parameters of the battery equivalent circuit. Therefore, it reduces the required number of tests to one. Extensive numerical studies on a 2 Ah Lithium-ion cell verify the effectiveness of the proposed method by achieving a RMS error less than one percent.