A Local Polynomial Approach to Nonparametric Estimation of the Best Linear Approximation of Lithium-Ion Battery From Multiple Datasets

Abstract

Battery short-term electrical impedance behaviour varies between linear, linear time-varying or nonlinear at different operating conditions. Data based electrical impedance modelling techniques often model the battery as a linear timeinvariant system at all operating conditions. In addition, these techniques require extensive and time consuming experimentation. Often due to sensor failures during experiments, constraints in data acquisition hardware, varying operating conditions and the slow dynamics of the battery, it is not always possible to acquire data in a single experiment. Hence multiple experiments must be performed. In this paper, a local polynomial approach is proposed to estimate nonparametrically the best linear approximation of the electrical impedance affected by varying levels of nonlinear distortion, from a series of input current and output voltage data sub-records of arbitrary length.

Incorporating Prior Knowledge in Local Parametric Modeling for Frequency Response Measurements: Applied to Thermal/Mechanical Systems

“…The result (8) reveals that transients lead to estimation errors of the true system G( k ). From ( 5) and (6), the system dynamics, i.e., pole locations, largely determine the transient contribution.…”

On the consistency and asymptotic normality of discrete-time LTI models identified from concatenated data sets

“…selection of operating condition or selection of informative data). Examples of such applications include: the linearmodeling of helicopter ight-mechanics from multiple short ight data [1]; the identication of linear-models for Lithium-Ion batteries and tokamaks with data from multiple experiments [2,3]; and the modeling of Linear Parameter-Varying systems from a local approach where, after operating condition selection, multiple records are available to identify a local Linear Time-Invariant (LTI) model [4].…”