State-of-Charge, State-of-Health and State-of-Power Estimation for Traction Batteries

Nowadays, Lithium-ion batteries (Li-Ion) are employed in many high-power applications such as energy storage systems and electric vehicles. Therefore, to ensure they safe operation, Battery Management System (BMS) monitor the battery modules, avoiding failures and ensuring its expected lifetime through the evaluation of three indexes: state-of-charge (SoC), state-of-health (SoH) and state-of-power (SoP). For this purpose, it is necessary to have an accurate estimation of these indexes, which degrade over time both in relation to the use of the battery and the external environmental parameters. In this paper, a new hybrid approach is proposed to estimate SoC, SoH and SoP based on series resistance <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$r_{s}$</tex> and incremental capacitance <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$c$</tex> values of Thevenin battery model (TBM) used to describe the dynamic behavior of the battery. A multi-layer-perceptron (MLP) neural network is designed to estimate <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$r_{{s}}$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$c$</tex> parameters of TBM. Finally, SoC, SoH and SoP are estimated by using a statistical approach. Numerical results and comparison analysis with the benchmark, show the effectiveness of the proposed method.