Dynamic temperature control in batteries is essential to avoid irreversible cell degradation and prevent thermal-runaway. The situation is exacerbated in e-mobility applications as the energy density and charge-discharge rates are increased to provide greater autonomy, shorter recharging times, and fast acceleration. To achieve homeostatic (i.e. self-regulated) isothermal operation of a battery stack, we plan to develop a BTMS prototype equipped with real dynamic control and sensing able to predict and regulate the temperature of the battery. An optimal BTMS will be enabled with the help of deep learning technology that uses the best available real-time probabilistic physics based models, experimental and sensor data, artificial intelligence, optimal control and monitoring to guarantee stable, efficient and safe battery performance.
PARTNERS INVOLVED
people involved
