Power Capability Estimation Accounting for Thermal and Electrical Constraints of Lithium-Ion Batteries.

Abstract

Lithium-ion (Li-ion) batteries have become one of the most critical components in vehicle electrification due to their high specific power and energy density. The performance and longevity of these batteries rely on constraining their operation such that voltage and temperature are regulated within prescribed intervals. Enforcement of constraints on the power capability is a viable solution to protect Li-ion batteries from overheating as well as over-charge/discharge. Moreover, the ability to estimate power capability is vital in formulating power management strategies that account for battery performance limitations while minimizing fuel consumption and emissions. To estimate power capability accounting for thermal and electrical constraints, the characterization of thermal and electrical system behavior is required. In the course of addressing this problem, first, a computationally efficient thermal model for a cylindrical battery is developed. The solution of the convective heat transfer problem is approximated by polynomials with identifiable parameters that have physical meaning. The parameterized thermal model is shown to accurately predict the measured core and surface temperatures. The model-based thermal estimation methodology is augmented for cases of unknown cooling conditions. The proposed method is shown with experimental data to accurately provide estimates of the core temperature even under faults in the cooling system. To jointly account for the thermal and electrical constraints, we utilize time scale separation, and propose a real-time implementable method to predict power capability of a Li-ion battery. The parameterized battery thermal model and estimation algorithms are integrated into a power management system for a series hybrid electric vehicle. An algorithm for sequential estimation of coupled model parameters and states is developed using sensitivity-based parameter grouping. The fully integrated co-simulation of the battery electro-thermal behavior and the on-line adaptive estimators reveal that the power management system can effectively determine power flow among hybrid powertrain components without violating operational constraints.