An Open-circuit-voltage Model of Lithium-ion Batteries for Effective Incremental Capacity Analysis

Open-Circuit-Voltage (OCV) is an essential part of battery models for state-of-charge (SOC) estimation. In this paper, we propose a new parametric OCV model, which considers the staging phenomenon during the lithium intercalation/deintercalation process. Results show that the new parametric model improves SOC estimation accuracy compared to other existing OCV models. Moreover, the model is shown to be suitable and effective for battery state-of-health monitoring. In particular, the new OCV model can be used for incremental capacity analysis (ICA), which reveals important information on the cell behavior associated with its electrochemical properties and aging status. INTRODUCTION With the widespread use of lithium-ion batteries for energy storage, the development of reliable and efficient battery management systems (BMS) has become a crucial task [1–3]. Two important functions of BMS are the state-of-charge (SOC) estimation and state-of-health (SOH) determination [4, 5]. SOC is commonly defined as “the percentage of the maximum possible charge that is present inside a rechargeable battery” (in this study, SOC is defined with respect to the current total capacity), and SOH is “a ‘measure’ that reflects the general condition of a battery and its ability to deliver the specified performance in comparison with a fresh battery” [6]. Typically, the quantitative definition of SOH is based either on the battery capacity or the internal resistance, depending on specific applications. The on-line estimation of battery SOC has been studied ex∗Address all correspondence to this author. tensively in literature (see Ref. [7] and references therein). Most of those methods, including the extended Kalman filter (EKF) approach [8–10], require an accurate open-circuit-voltage (OCV) model which relates OCV to SOC [7, 11, 12]. The OCV-SOC function is implemented either as a look-up table or an analytical expression, while the latter has several advantages including computational efficiency (since no interpolation is needed) and ease for analysis [13]. Several analytical OCV models proposed in the literature are summarized in Ref. [13]. Those are phenomenological models built with curve fitting without considering the complex battery physical behavior during the lithium-ion intercalaction/deintercalation process [14,15]. However, as reported in our previous work [5], the OCV data obtained from the galvanostatic charging/discharging of batteries at low C rate displays voltage plateaus and transitions (please see the voltage curve plotted in Fig. 1) that correspond to the staging phenomenon at the graphite anode [5, 16–19]. Without proper parametrization, this critical phenomenon is usually lost when the OCV data is fitted with those existing models. At the same time, because of the wide flat region on the OCV-SOC curve, a small mismatch in OCV fitting may cause a large deviation when used for SOC estimation. Moreover, OCV data often reflect battery aging and performance degradation [20]. Using the so-called incremental capacity analysis (ICA) technique [16, 21], which differentiates the battery charged capacity (Q) with respect to the terminal voltage (V ) and transforms voltage plateaus on theV -Q (voltage-charged versus capacity) curve into clearly identifiable dQ/dV peaks on the incremental capacity (IC) curve (see Fig. 1), gradual changes in cell behavior can be detected, based on life cycle test data, 1 Copyright © 2013 by ASME Proceedings of the ASME 2013 Dynamic Systems and Control Conference DSCC2013 October 21-23, 2013, Palo Alto, California, USA