State of health estimation of Lithium-ion batteries using operational data based on approximate weighted total least squares method

Conference: NEIS 2023 - Conference on Sustainable Energy Supply and Energy Storage Systems
09/04/2023 at Hamburg, Germany

Proceedings: NEIS 2023

Pages: 8Language: englishTyp: PDF

Authors:
Vaidya, Aditya Madhav (Vattenfall Solar GmbH, Hamburg, Germany & HAW Hamburg Fakultät Life Sciences, Hamburg, Germany)
Alhaider, Firas (Vattenfall Solar GmbH, Hamburg, Germany & Helmut-Schmidt-Universität, Fakultät für Elektrotechnik, Hamburg, Germany)

Abstract:
With the increasing demand for renewable energy, the importance of energy storage is also increased. Lithium-ion bat-teries are widely used for energy storage applications. State of charge (SoC) and State of health (SoH) are critical param-eters of battery management systems and must be determined precisely. These are the parameters which are difficult to measure directly, hence indirect means must be adopted. This paper is a continuation of the research work conducted by the author in [1]. In his work he had pointed out the limitations of the model and the lack of recursive algorithm. This paper demonstrates a recursive and implementation of fading memory to place more confidence on the recent measure-ments. The objective of this technical paper is to estimate the Capacity and State of Health (SoH) of operational battery assets by utilizing solely two parameters, voltage and current. The recursive algorithm with fading memory gave better results than what was reported in [1]. For test cell data it showed estimation error well below 0.7% and for real-life battery data it produced realistic estimates.