BESS modeling: best available approaches for modeling from measurements

Abstract

The rapid deployment of lithium-ion Battery Energy Storage Systems (BESS) in modern power systems has increased the need for accurate, computationally efficient, and application-oriented battery models to support design, operation, control, and lifetime management. This paper presents a comprehensive review of lithium-ion battery modeling approaches developed and validated from measurement data, including electrochemical, equivalent-circuit, data-driven, empirical, and hybrid frameworks. The core contribution of the review is a structured classification framework centered on the data landscape, which explicitly links data availability and measurement level to application requirements and model selection. Four main application domains are considered: Battery Management Systems, Energy Management Systems, degradation and lifetime prediction, and safety analysis. For each domain, the review examines the relevant data sources, measurement requirements, parameter-identification methods, computational constraints, and the advantages and limitations of the different modeling frameworks. A comparative analysis of the reviewed literature is then conducted to identify recurring relationships among application objectives, available data, model complexity, and implementation requirements. The findings show that the most suitable modeling framework is not determined by fidelity alone, but by its consistency with the intended use, the available measurements, and the computational resources of the application. Building on these findings, practical model-selection guidelines are proposed to support the choice of appropriate modeling strategies for real-world BESS. Finally, the review highlights current research gaps and future directions toward more reliable, transferable, and practically deployable BESS modeling frameworks.