Algebraic method for parameter identification of circuit models for batteries under non-zero initial condition

Effects of rest time on discharge response and equivalent circuit model for a lead-acid battery

“…In this section, we give a brief review of the algebraic method developed in [26]. A second-order Thevenin's equivalent circuit model is depicted in Fig.…”

“…Third or higher order models may further reduce the root-mean-squareerror but the parameters would be too sensitive to noises and measurement errors. As shown in [26], the parameters for a third-order model are very sensitive to voltage perturbations. In [19], it was demonstrated that two third order models can produce very close voltage responses even though their parameters are very different.…”

“…Our recent work [26] was motivated to reduce the rest time between two discharging tests by eliminating the requirement of 0 initial conditions. In particular, we developed an algebraic method for extracting the M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT parameters together with the initial conditions.…”

“…This fact was mathematically proven in [26]. Thus it is impossible to extract the circuit parameters under non-zero initial conditions with a constant load current.…”

“…Our study is enabled by a recently developed method for parameter identification of Thevenin's equivalent circuit model under non-zero initial conditions [26]. The elimination of the assumption on 0 initial condition not only reduces the rest time to almost 0 and drastically accelerate the process for battery modeling, but also opens a door to the understanding of the effects of rest time on the subsequent discharge response.…”