A Model Reference Adaptive Controller for an Electric Motor Thermal Management System in Autonomous Vehicles

Technological advancements and growth in electric motors and battery packs enable vehicle propulsion electrifications, which minimize the need for fossil fuel consumption. The mobility shift to electric motors creates a demand for an efficient electric motor thermal management system that can accommodate heat dissipation needs with minimum power requirements and noise generation. This study proposes an intelligent hybrid cooling system that includes a gravity-aided passive cooling solution coupled with a smart supplementary liquid cooling system. The active cooling system contains a radiator, heat sink, variable frequency drive, alternating current (AC) fan, direct current (DC) pump, and real-time controller. A complete nonlinear mathematical model is developed using a lumped parameter approach to estimate the optimum fan and pump operations at each control interval. Four different control strategies, including nonlinear model predictive controller, classical proportional-integral (PI) control, sliding mode control (SMC), and stateflow (SF), are developed, and their performance is compared. The experimental results demonstrate that the nonlinear model predictive control (NMPC) method is the most effective strategy, which reduces the cooling system fan power consumption by 73% for only a 5% increase in the pump power usage compared to classical PI control for a specific 60-minute driving cycle.