Temperature Control Valve PID Controller Modeling and Correlation

Proportional Integral Derivative (PID) controllers are used frequently in thermal systems to manage the temperature of components and environments. Tuning a PID controller for a Thermal Control System (TCS) can be difficult without having the complete system assembled, as well as access to a test environment that can accommodate the wide range of environmental conditions. Spacecraft pumped fluid loop thermal control systems often utilize a radiator and a temperature control valve (TCV) to manage radiator flow and reject heat. The TCV proportionally sends coolant to either the radiator or the bypass leg to maintain the desired temperature setpoint. The TCV position is manipulated through a motor that is operated by a PID controller that monitors the temperature of the coolant. Simulation can accelerate controller tuning and augment environmental testing of the complete system assembly. Thermal Desktop FloCAD was used to model a wide range of physical parameters such as flow geometry and drive train backlash as well as vehicle software limitations such as command latency. For the presented case, test data used with simple thermal modeling constructs (e.g. lump capacitances) allowed a FloCAD PID Controller model to be correlated to tested performance with relatively good accuracy. After model correlation was achieved, controller parameter sensitivity was then investigated to aid in troubleshooting issues while on mission or improving future systems. This paper goes through the process of modeling known system constraints, correlating unknown constraints with test data, and studying the sensitivity of control parameter inputs.