Abstract
The integration of renewable energy sources is a challenge for the energy industry of the future. As an additional buffer between electricity supply and demand, batteries are an increasingly important candidate. This is facilitated by their rapidly decreasing cost and flexibility.They will play an important role in supporting grid stability and flexibility. In addition to power generation, road transport is becoming increasingly electrified: Sales of electric cars are growing rapidly, which means that charging them can put a significant strain on power grids.As a result, there is a need for efficient battery models in terms of speed and accuracy. These can be used for the development, testing and simulation of advanced charging algorithms for electric vehicles, for the planning of modern energy grids that include complete battery storage systems, and for the characterization of newly developed batteries and the prediction of their behavior in different scenarios. This work is intended to contribute to the solution of these problems.The following work presents a complete simulation solution of a battery system and the basics of an electric vehicle emulator based on physical battery models. In particular, simulation methods are developed to simulate fast charging algorithms for electric vehicles. First, the thermodynamic principles of electrochemical cells that form the basis of these technologies are summarized. This is followed by an explanation of the processes involved in battery degradation.The next theoretical point of the thesis is an overview of the most important international standards for charging electric vehicles. This is followed by a summary of the basics of classical and new charging algorithms for electric vehicles. Modern battery technologies in electric vehicles are then presented. The last part of the second chapter is dedicated to the task of presenting and evaluating the best known modeling approaches for Li-ion batteries.Based on the theoretical foundations of Chapter 2, the concept of the simulation approaches is developed step by step in Chapter 3. This starts with a precise description of the model used in this thesis. A method to parameterize this model is developed. Then a new method based on adaptive polynomial interpolation is presented to establish mathematically smooth functional dependencies between the state of charge of the battery and the values of the components of its equivalent circuit diagram. After establishing an accurate model for simulations of the transient behavior of individual cells, approaches for scaling cells at the battery pack level are presented. These exploit both the linearity of the resistive-capacitive equivalent circuit diagrams and the differential equations that describe them. Since it is now possible to simulate not only individual battery cells but also entire battery packs, the conditions specified by the (fast) charging algorithms are integrated into the simulation of the charging behavior itself. Several commonly used charging algorithms are implemented.A comparison of measured and simulated battery cell voltage profiles shows that the developed models deliver very accurate results as soon as the corresponding input parameters are available. A comparison with traditional methods (non-adaptive polynomial fitting and lookup tables) is also made and it is shown that the models developed in this thesis are superior to the conventional methods. In the second part of Chapter 4, the models for simulating the fast charging behavior of electric vehicles are validated. For this purpose, publicly available charging curves are used and re-simulated. The simulation data are post-processed and the resulting data are compared with the manufacturer's specifications for the respective vehicles to provide a further validation method.At the end of the thesis, further recommendations and possible applications for the developed tool are given.
Originalsprache | Englisch |
---|---|
Gradverleihende Hochschule |
|
Betreuer/-in / Berater/-in |
|
Datum der Bewilligung | 3 Juli 2024 |
Erscheinungsort | 2024 |
Publikationsstatus | Veröffentlicht - 2024 |
Research Field
- Hybrid Power Plants