New hybrid controller cuts EV heat pump energy use by 28%

Thermal management is a major energy drain in battery electric vehicles, especially in cold climates where heating the cabin and battery can reduce range by 30–40%. Prashant Lokur and Nikolce Murgovski from Chalmers University of Technology present a novel hybrid control framework that optimally coordinates the compressor, coolant pumps, and cabin blower across the refrigerant, coolant, and air loops. The framework uses a rule-based supervisory layer to handle discrete system configurations (e.g., switching between heating modes) and a continuous nonlinear model predictive controller (NMPC) that minimizes thermal energy consumption over a finite horizon. The control-oriented model captures the dominant dynamics of the cabin, refrigerant loop, reconfigurable coolant circuits, and key thermal masses (battery, motor, inverter). The terminal cost is computed by linearizing the system about a quasi-steady point and solving the discrete-time algebraic Riccati equation, ensuring well-conditioned optimization across varying conditions.

The framework was validated against the high-fidelity MathWorks Simscape "Electric Vehicle Thermal Management with Heat Pump" model. It achieved a mean absolute temperature prediction error below 1.8°C for battery, motor, and cabin air temperatures, while reducing simulation time by approximately 85%. Under cold-climate extended driving scenarios, the proposed controller consistently reduced thermal energy consumption by 20–28% compared to the built-in rule-based controller. This translates to meaningful range extension for EVs in winter conditions without sacrificing comfort. The complete implementation, built using the open-source CasADi optimization framework, is publicly available on GitHub, enabling replication and further development by the automotive industry and research community.