Thesis title: Modelling and Real-Time Predictive Control for Hybrid Cooling of Lithium-Ion Batteries

Abstract:

Surface cooling offers a large heat-transfer area, while tab cooling can reduce internal temperature gradients by exploiting the high thermal conductivity of battery current collectors. This thesis investigates how these complementary cooling approaches can be integrated through modelling and predictive control.

Reduced-order two-dimensional thermal models are developed to reproduce the spatial temperature field at low computational cost. Building on these models, a real-time model predictive control framework dynamically distributes coolant between the battery tabs and lateral surface. The proposed approach closely reproduces nonlinear model predictive control while reducing the mean computation time from several seconds to approximately 19 ms. The framework is further extended to account for battery degradation and spatial temperature non-uniformity, reducing degradation-model computation by more than 20 times. Overall, integrated tab-surface cooling achieves the best balance among temperature regulation, thermal uniformity, battery degradation, and coolant use.

For more information, visit Chalmers University of Technology calendar