Robust parameter estimation approach of Lithium‐ion batteries employing bald eagle search algorithm

Abstract

Summary Lithium‐ion batteries play an essential role in sustainable power systems as energy storage systems. They are employed for electrical power storage and balancing power in multisources power systems. However, due to the degradation phenomena, these systems are more exposed to parametric variation. For this reason, the identification of the battery parameters remains crucial to enhance the power system performance and extend the battery lifespan. A robust parameter identification strategy based on the bald eagle search algorithm (BES) is proposed in this research work. This article proposed a robust estimation approach that accurately estimates the battery parameters based on the BES optimizer. The proposed approach has been evaluated for two distinct batteries to confirm its superiority. A comparison with recent optimization methods has been performed to validate the performance of the proposed strategy. These algorithms include marine predator algorithm, COOT algorithm, artificial eco‐system optimizer, and other previous algorithms. The obtained results affirmed the reliability of the proposed BES‐based strategy to estimate the battery equivalent circuit variables with superior accuracy compared with other methods. It provided the smallest fitness function value for the first battery with 7.06 × 10−4 and an excellent SD value of 1.877 × 10−8. The proposed strategy achieved 100% efficiency in terms of optimization efficiency. The second battery has been tested with ArtUban and New European Driving Cycle (NEDC) profiles to confirm the performance of BES. The results confirm the superior performance for the two cases. The BES provides the smallest fitness values (0.0860 for ArtUrban and 0.1222 for NEDC) and lower identification total error (2.294 for ArtUrban and 2.6737 for NEDC).

Optimal parameters identification strategy of a lead acid battery model for photovoltaic applications

“…In this work, the study is focused on the extraction strategy of LaB parameters. Battery models require the knowledge of appropriate parameters to describe charge and discharge processes 9,10 . The objective is to build a workspace to analyze LaB system storage problems, from its design to its starting and operating.…”

“…Battery models require the knowledge of appropriate parameters to describe charge and discharge processes. 9,10 The objective is to build a workspace to analyze LaB system storage problems, from its design to its starting and operating. The identification techniques of this study are based on meta-heuristic optimization algorithms (MOAs).…”

Simple parameterization method using differential evolution algorithm to estimate state of charge for Li‐ion batteries and packs for electric vehicles

“…To avoid the convergence time, a separation between state and parameter estimation could be synergistic, in this way, there are some options with different methodologies [15][16][17] but they are specially implemented for a type of chemistry or modeling. Another proposed solution is the use of optimization methods to search the optimal parameters, 18,19 including some toolboxes of the MATLAB software 20,21 used to obtain the parameters. The main disadvantage from the aforementioned options is the use of built-in functions that could limit the results, and the obvious dependence of commercial software.…”