Scaling Electrochemical Battery Models for Time-Accelerated and Size-Scaled Experiments on Test-Benches

Abstract

This paper presents a dimensional-analysis supported scaling procedure applied to a mathematical model of electrochemical batteries. The main objective of this research is to allow for laboratory size-scaled and time-compressed experimental analysis of processes involving large physical magnitudes and evolving over long time spans. These situations are of interest when considering the sizing of battery packs and other components of energy systems, particularly smart grids, and further systems where battery storage is relevant, like hybrid vehicles and other standalone systems, as well as deciding management strategies on them. Voltage-, current- and time-scaled models preserving the dynamic evolution of a group of relevant physical magnitudes are presented. These models have been validated through simulation and physical experiments on a test-bench designed and constructed on purpose. The physical implementation of the scaled models is not possible in the cases where some of the scaled model parameters cannot be met using real batteries. But, as the mathematical construction of the scaled models is always possible, this problem can be circumvented with a Hardware-in-the-loop approach: the scaled battery is numerically emulated on a programmable and controllable power source/sink system, which is run in real-time embedded in the test-bench representing the whole system under study.