The necessity for research into the effects of temperature on lithium-ion batteries and battery packs is obvious and important in order to develop electric vehicles that are widely adopted by the public. Models that quickly and accurately forecast the temperature and voltage depending on operational parameters can avoid thermal runaway, increase charging speed, prevent lithium plating, and increase cycle life.
The work consists of a thorough investigation of the temperature effect on various battery parameters such as state of health and internal resistance.
The effect of temperature on overpotential and current density are also modeled. The thesis is divided into four parts:
Part 1:
This section introduces two distinct models: an internal resistance (Rint) model and a physiological-chemical partial differential model. The Rint model models as a function of state of charge (SOC), and C-rate. Simulation results are compared to experimental data at various C-rates (1-4C) at 22°C.
Simulation findings indicate accurate temperature prediction using a simple Rint model. The reduced physiochemical model with only three partial differential equations achieves comparable accuracy to the Rint model. The Rint model accurately predicts battery internal resistance using a Pearson curve and hyperbolic sine function.
Part 2:
This section show cases three electrothermal equivalent circuit models with multiple input parameters. The model allows us to estimate parameters like internal impedance using practical inputs. The study simulates the internal impedance resistance of a LiFePO4 battery at various ambient temperatures
(5-45 °C), discharge rates (1-3C), and SOHs (65-90%). Three thermal models incorporated the internal resistance surface model. The first two thermal models were 0D and did not account for the battery's thermal conductivity.
The first model assumed simple heating from internal resistance and convective energy loss, while the second incorporated the Bernardi Equation Reversible heat term. The third model was a 2D model that retained the earlier heat source terms while adding a tab junction heating source term.
The 2D model was solved with a basic Euler approach and finite center difference method.
Part 3:
This research reported laboratory data and modeling results for C-rates of 1-4C at an ambient temperature of approximately 23°C. Experiments were carried out at continuous current discharge. Simulation results indicate that the cathode generates more heat than the anode, with electrolyte resistance being the dominant source of heat. Battery temperature was highest near tabs and within the battery’s internal space. The simulation of lithium concentration in the battery revealed that the anode had a more uniform concentration than the cathode. These findings can aid in the precise thermal design and control of Li-ion batteries.
Part 4:
The experimental setup consisted of 7 Panasonic NCA cells connected in parallel, with each cell rated at 3.2Ah capacity. Individual cell capacities were measured and averaged, and the experimentally determined value was 3.11Ah. The batteries were allowed to equilibrate to a steady voltage at the end of discharge. The battery's thermal behavior was measured at six different discharge rates (constant current): 0.5C to 1.75C.
Supervisor: Professor Fowler