Aristotle University Of Thessaloniki (AUTH)
Dynamic analysis and simulation of lithium-ion battery energy systems
Abstract
dc:descriptionThe decarbonisation of energy systems and the electrification of transport are central components of the global response to climate change. Lithium-ion batteries, due to their high energy density, efficiency, and rapid response characteristics, have become a cornerstone technology across electric mobility, buildings, and power systems. As deployment expands, there is an increasing need for accurate, scalable, and transparent modelling tools capable of capturing battery behaviour under realistic and dynamically varying operating conditions. This dissertation addresses that need through the development and validation of LIBSystems, a flexible multi-domain modelling framework for simulating lithium-ion battery systems across multiple application scales. LIBSystems is a modular library developed in Modelica, designed to support physically interpretable, equation-based modelling of lithium-ion batteries from the single-cell level to full energy systems. It includes equivalent-circuit cell models, pack-level aggregation, electro-thermal coupling, and a semi-empirical ageing formulation. The library also incorporates auxiliary components such as vehicle powertrains, bidirectional chargers, and grid-connected inverters, enabling seamless integration of batteries within broader energy systems. The design prioritises transparency, adaptability, and interoperability, avoiding black-box structures and supporting connectivity with external tools through FMI and API interfaces. The library is complemented by a wide collection of open parameter datasets covering commercial lithium-ion chemistries and capacities, ensuring representativeness and reproducibility. The dissertation begins by establishing the policy, technological, and market context that has driven rapid lithium-ion adoption. A concise theoretical background follows, covering electrochemical fundamentals, cell-level quantities, and degradation mechanisms. This is complemented by a review of modelling approaches, including physics-based electrochemical models, equivalent-circuit representations, and data-driven techniques. The trade-offs between model fidelity, computational tractability, and scalability highlight the need for a unified, modular, physics-consistent modelling environment capable of serving both detailed analysis and system-level applications. A rigorous verification and validation workflow was implemented to assess LIBSystems across multiple time and spatial scales. At the cell level, short-term transient accuracy and long-term consistency were evaluated under standard driving cycles (US06, FTP-75) using both experimental data and PyBaMM physics-based references. Hybrid Pulse Power Characterisation tests further validated voltage response fidelity. At the pack level, aggregation behaviour and scalability were verified against liionpack-generated multi-cell voltage profiles. The ageing module was calibrated using a dataset of 228 cells, capturing both calendar and cycling degradation, enabling predictive simulation of long-term capacity fade. At the system level, a complete electric vehicle model reproduced voltage, power, and state-of-charge trajectories of a validated MATLAB Simscape benchmark, confirming consistency from component to full-system scale. The modelling framework was subsequently deployed in four applications. In a residential PV–ORC–battery microgrid, replacing the lead-acid unit with a lithium-ion system increased PV utilisation by 4.5% and raised the ratio of injected to nominal energy capacity by 10% due to deeper cycling capability. In the study of solar-powered electric vehicles, annual grid energy consumption per 100 km in Berlin decreased by 29% with rooftop-mounted PV and by 45% when charging through a residential PV-battery system; in Los Angeles, the reductions were 44% and 74% respectively. Daily solar-driving range increased on average by 13.2 km in Berlin and 19.7 km in Los Angeles, demonstrating strong localisation effects driven by solar resource availability. In distribution grids equipped with vehicle-to-grid chargers, dynamic simulations using historical load data confirmed the technical feasibility of the examined V2G topology and scheduling, showing that coordinated charging and discharging can be achieved without compromising grid stability. In the Terceira Island power system, a 10.5 MW / 15 MWh lithium-ion BESS increased the frequency nadir by 14% under 60% RES penetration and by 17% under 80% RES penetration, while recovery time in the 60% RES scenario decreased from 13 s to 10 s, highlighting the contribution of fast-response storage in low-inertia islanded conditions. Overall, the dissertation demonstrates that LIBSystems constitutes a robust and transparent ecosystem linking lithium-ion cell operation with large-scale energy-system simulation. The framework brings together modular battery modelling, multi-scale validation and system-level deployment. It provides a foundation for future extensions involving physics-based ageing mechanisms, advanced thermal and power electronics submodels, hardware-in-the-loop implementations and advanced control strategies, supporting the design and optimisation of battery energy systems.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Rotas, Renos Panagiotis
- Ρώτας, Ρένος-Παναγιώτης
Subjects
dc:subject × 14- Συσσωρευτές ιόντων Λιθίου
- Δυναμική προσομοίωση
- Ολοκληρωμένα ενεργειακά συστήματα
- Αποθήκευση ενέργειας
- Lithium-ion batteries
- Dynamic simulation
- Integrated energy systems
- Energy storage
- Επιστήμη Μηχανολόγου Μηχανικού
- Επιστήμες Μηχανικού και Τεχνολογία
- Μηχανολογία
- Mechanical Engineering
- Engineering and Technology
- Mechanics
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/60514