Development and validation of a parameter estimation methodology for two different lithium-ion batteries to optimize their performance and life cycle

Abstract

This paper presents four procedures developed to analyse the dependence of the discharge curve of a lithium battery on discharge current and working temperature. For this work, two models of lithium batteries have been used, whose discharge curves have different shapes. The first one, the shape of the curve describes an almost horizontal line in most of it (type 1), while in the other one, the shape of the curve describes a negative slope in most of it (type 2). Two of the developed procedures are used for discharge curves with different currents but constant temperature and the other two developed procedures are used for discharge curves with different temperatures but constant current. The information for the development of the simulations is obtained from the datasheet of the analysed lithium batteries.

Estimation of cycling aging of lithium-ion batteries for photovoltaic applications

“…The present work models the electrical characteristic of a lithium battery using equation (1) for discharge curve and equation ( 2) for charge curve [1]. The dependence of the model parameters with respect to temperature and current have been established in [2].…”

“…The extraction of the model parameters will be based on the discharge curves. The parameter 𝐸 𝑜 was estimated as in [2] and the parameters 𝐴 and 𝐵 are estimated as in [5]. The terms 𝑅 𝑇 𝑟𝑒𝑓 , 𝜕𝐸 𝜕𝑇 ⁄ and 𝛼, that appear in [2] and which will also be used for this work, were estimated as in [2].…”

“…The parameter 𝐸 𝑜 was estimated as in [2] and the parameters 𝐴 and 𝐵 are estimated as in [5]. The terms 𝑅 𝑇 𝑟𝑒𝑓 , 𝜕𝐸 𝜕𝑇 ⁄ and 𝛼, that appear in [2] and which will also be used for this work, were estimated as in [2]. It has been assumed that 𝐾 𝑖 𝑟𝑒𝑓 = 𝐾 𝑇 𝑟𝑒𝑓 and 𝑄 𝑖 𝑟𝑒𝑓 = 𝑄 𝑇 𝑟𝑒𝑓 , these terms also appear in [2] and are indicated in this work as 𝐾 𝑟 and 𝑄 𝑟 , respectively.…”

“…The terms 𝑅 𝑇 𝑟𝑒𝑓 , 𝜕𝐸 𝜕𝑇 ⁄ and 𝛼, that appear in [2] and which will also be used for this work, were estimated as in [2]. It has been assumed that 𝐾 𝑖 𝑟𝑒𝑓 = 𝐾 𝑇 𝑟𝑒𝑓 and 𝑄 𝑖 𝑟𝑒𝑓 = 𝑄 𝑇 𝑟𝑒𝑓 , these terms also appear in [2] and are indicated in this work as 𝐾 𝑟 and 𝑄 𝑟 , respectively. Then, the parameters 𝐾 (𝑖,𝑇) and 𝑄 (𝑖,𝑇) are estimated using equation ( 7) and (8).…”