Application of Model Predictive Control to Cabin Climate Control Leading to Increased Electric Vehicle Range 2023-01-0137

For electric vehicles (EVs), driving range is one of the major concerns for wider customer acceptance and the cabin climate system represents the most significant auxiliary load for battery consumption. Unlike internally combustion engine (ICE) vehicles, EVs cannot utilize the waste heat from an engine to heat the cabin through the heating, ventilation and air conditioning (HVAC) system. Instead, EVs use battery energy for cabin heating, this reduces the driving range.

To mitigate this situation, one of the most promising solutions is to optimize the recirculation of cabin air, to minimize the energy consumed by heating the cold ambient air through the HVAC system, whilst maintaining the same level of cabin comfort. However, the development of this controller is challenging, due to the coupled, nonlinear and multi-input multi-output nature of the HVAC and thermal systems. Furthermore, the controller must satisfy different control requirements by leveraging multiple control actuators whilst simultaneously respecting multiple control and system constraints. A Model Predictive Controller has been used to reduce the control complexity of a conventional controller that requires many heuristic control laws and significant calibration effort. The control solution has been realised for an industrialized application, respecting typical embedded control hardware solutions, minimizing computational effort and memory requirements.

This paper introduces a development approach utilizing Adaptive Model Predictive Control (AMPC) to address the challenges above, leading to the demonstration in a vehicle (Jaguar I-PACE) over the UDDS cycle on a climatic chassis dynamometer. The structured development approach improves the efficiency and flexibility of setting-up advanced control solutions.