Real-time discrete suboptimal control for systems with input and state delays: Experimental tests on a dehydration...

Area of Science:

• Control Systems Engineering
• Applied Mathematics
• Chemical Process Control

Background:

• Affine nonlinear discrete systems often exhibit complex dynamics due to state and input delays.
• Traditional control strategies may struggle with the computational complexity of optimal control for such systems.
• Finite horizon control is crucial for practical applications requiring timely responses.

Purpose of the Study:

• To develop a suboptimal control strategy for affine nonlinear discrete systems with state and input delays.
• To address the computational challenges of the Dynamic Programming Approach by proposing a numerical approximation.
• To validate the proposed control strategy through experimental testing and comparison with existing methods.

Main Methods:

• Utilized the Dynamic Programming Approach to derive a suboptimal control sequence.
• Implemented a numerical approximation for the Bellman functional to avoid complex computations.
• Conducted experimental tests on a dehydration process for validation.

Main Results:

• The proposed suboptimal control strategy demonstrated good performance and behavior in the dehydration process.
• The numerical approximation of the Bellman functional proved feasible and effective.
• Comparison with a Proportional Integral Derivative (PID) controller tuned via Ziegler-Nichols highlighted the benefits of the proposed strategy.

Conclusions:

• The developed suboptimal control strategy offers a viable and effective solution for systems with state and input delays.
• Numerical approximation of the Bellman functional is a practical approach to overcome computational hurdles in Dynamic Programming.
• The strategy shows superior performance compared to conventional PID control in the tested dehydration process.