How to parameterise an equivalent-circuit empirical battery model from time-domain data

Area of Science:

• Electrochemistry
• Computational Modeling

Background:

• Empirical battery models are crucial for applications like pack design and charge estimation.
• Accurate model parameters are essential, but parameter fitting involves often-overlooked subtleties.
• Existing literature frequently misses key aspects of robust parameterization.

Purpose of the Study:

• To present a best-practice guide for empirical battery model parameterization.
• To address subtleties in parameter fitting often missed in the literature.
• To provide a generalizable method for users to select appropriate experimental and fitting setups.

Main Methods:

• Designing appropriate experiments for data acquisition.
• Utilizing numerical optimizers for fitting model parameters to experimental data.
• Developing parameter functions through interpolation of fitted parameters.
• Employing a multi-step parameterization process with gradually increasing model complexity.
• Prioritizing timescales over capacitances for enhanced numerical stability.

Main Results:

• A comprehensive guide for empirical battery model parameterization is presented.
• Novel insights include the use of timescales for numerical stability and a multi-step fitting process.
• A combined numerical and experimental design approach is detailed for optimal parameter estimation.
• Provided scripts, data, and fitted parameters facilitate method testing and re-use.

Conclusions:

• The presented method offers a reliable and generalizable approach to empirical battery model parameterization.
• Adherence to best practices ensures accurate parameter estimation for diverse battery modeling applications.
• The provided resources support the adoption and advancement of battery modeling techniques.