Optimal control analysis and Practical NMPC applied to refrigeration systems.

Area of Science:

• Thermodynamics and Refrigeration Systems
• Control Theory and Engineering

Background:

• Mechanical compression refrigeration systems are crucial for cooling applications.
• Optimizing their efficiency while meeting cooling demands is a significant engineering challenge.
• Existing control strategies often struggle to achieve theoretical optimal performance.

Purpose of the Study:

• To develop and analyze a reduced-order state-space model for optimal control of canonical refrigeration cycles.
• To investigate the controllability of the proposed model in achieving optimal operational states.
• To compare the performance of a practical Nonlinear Model Predictive Control (NMPC) strategy against a known feedback-plus-feedforward approach.

Main Methods:

• A reduced-order state-space model based on the moving boundary approach was developed for the canonical refrigeration cycle.
• Controllability analysis was performed to understand the limitations in achieving optimal cycle variables.
• Optimization simulations were conducted to evaluate control strategies under varying cooling demands.

Main Results:

• The optimal refrigeration cycle is defined by three variables but only two control inputs are available, posing a controllability challenge.
• Optimization simulations revealed that minimum superheating does not always lead to optimal cycles for varying cooling demands.
• Both practical NMPC and a feedback-plus-feedforward strategy encountered difficulties in reaching the optimal cycle.

Conclusions:

• The proposed reduced-order model highlights inherent controllability limitations in achieving optimal efficiency for mechanical compression refrigeration systems.
• Control strategies, including NMPC, may struggle to reach theoretical optimal performance due to these limitations.
• Further research into advanced control techniques or system modifications may be necessary to overcome these challenges.