Model predictive control of temperature-sensitive and transient loads in aircraft vapor compression systems

Abstract

Challenges for next generation aircraft thermal management using vapor compression systems (VCS) include larger transient thermal loads, tighter temperature constraints, and increasingly strict power and sizing requirements. Conventional control approaches, such as proportional-integral (PI) control, are well-suited for steady-state operation without system constraints. In this work, we investigate the ability of model predictive control to enable larger transient thermal loads subject to tighter temperature limits through anticipation and load-shedding. Our results include (1) a linear model predictive control validated on a nonlinear model of the vapor compression system, and (2) direct comparisons to state-of-the-art PI control for stressing scenarios.