Electrode State of Health Estimation for Lithium Ion Batteries Considering Half-cell Potential Change Due to Aging

Abstract

Estimation of electrode state of health (eSOH) is essential to understanding battery degradation status in detail. This is accomplished by considering electrode capacity and a utilization range as eSOH parameters. In this paper, we propose a novel combination of two eSOH estimation approaches (i.e. voltage fitting and differential voltage analysis). By utilizing peak information in the differential voltage curve, the proposed method can separate individual electrode's contributions from the full-cell voltage. This separation allows the proposed method to identify changes in the positive electrode half-cell potential curve and to calibrate the aged half-cell potential function by refitting the coefficient of the basis functions. A comparison between the conventional voltage fitting method and the proposed method for an aged NMC/graphite cell shows that the proposed method can significantly reduce the voltage fit error while matching the location of distinct peaks in the differential voltage curve that the conventional method fails to do.

Predicting the impact of formation protocols on battery lifetime immediately after manufacturing

“…We performed a voltage fitting analysis 44,[52][53][54][55] to confirm that the main failure mode in our cells is the loss of lithium inventory (LLI) over cycle life (Figures S18 and S19). We found that LLI can fully account for the thermodynamic (i.e., C/20) cell capacity loss over life.…”

A Comprehensive Review of Available Battery Datasets, RUL Prediction Approaches, and Advanced Battery Management

“…The electrode state of health is achieved by utilization range as an eSOH parameter and considering electrode capacity. However, the electrode-specific state of health inside a cell has been given less attention [115]. Electrode-specific state of health gives elaborated information on the degradation status of the battery.…”

“…Voltage fitting finds a parameter set with optimized algorithms. The parameter set gives the best fit between measured data and model prediction for the voltage curve [115]. Using a genetic algorithm, a battery model was proposed with electrode parameters by X. Han et al [116], and that was identified by fitting the voltage curve.…”

Differential Expansion and Voltage Model for Li-ion Batteries at Practical Charging Rates

“…wheret 0 + = t 0 + (c 0 e ) andD e = D e (c 0 e ) with the same boundary conditions as the ones for Eq. (8). Similarly for solving the liquid-phase Ohm's law, Eq.…”

“…For example, it has been shown that without knowledge of the internal state of the cell we could charge the cell blindly with respect to lithium plating, even in cases of moderate capacity fade [2]. To estimate the degradation mechanisms many studies have utilized the voltage [3,4,5,6,2,7], the differential voltage and phase transitions [8]. All these methods require full range voltage data from the open circuit (OC) state or pseudo-OC state, at a very low current rate.…”