Advanced nonlinear model predictive control for electric vehicle battery thermal management under the Addis Ababa Bishoftu driving cycle

Robust battery thermal management (BTM) is a critical enabler of electric vehicle (EV) safety, longevity, and performance, particularly under dynamic and geographically complex driving conditions. This study introduces a nonlinear model predictive control (NMPC) framework for lithium-ion battery thermal regulation, assessed through a simulation-based evaluation employing a physics-informed, reduced-order lumped-parameter thermal model under the real-world Addis Ababa–Bishoftu driving cycle a 45 km route characterized by steep elevation gradients, variable ambient temperatures, and mixed traffic patterns. The thermal model was implemented and integrated within a LabVIEW–MATLAB co-simulation platform to replicate electrothermal behavior with spatial and temporal precision. Compared to conventional linear MPC, the NMPC approach reduced peak inter-cell temperature differentials by over 60% and achieved a 25–30% reduction in coolant air velocity, without violating safety constraints. Under both regional and standard US06 driving cycles, the NMPC controller demonstrated superior tracking accuracy, energy efficiency, and real-time computational feasibility (< 150 ms per update). These results underscore the advantages of nonlinear predictive control in capturing complex thermodynamic interactions, enabling context-aware thermal regulation that is both safe and energy-efficient. The proposed framework provides a simulation-level proof of concept for intelligent BTM strategies in EVs operating in topographically and climatically diverse regions, with experimental validation identified as essential future work to confirm real-world applicability.