Swarm optimization approach to design PID controller for artificially ventilated human respiratory system

Area of Science:

• Biomedical Engineering
• Control Systems
• Computational Intelligence

Background:

• Artificial ventilation systems are crucial for patients with respiratory issues, requiring precise control of airway pressure.
• Optimizing controller parameters for medical ventilators is complex due to patient-specific factors.
• Computer-aided algorithms enhance the performance and control of artificial ventilation.

Purpose of the Study:

• To design and verify a swarm optimization-based controller for pressure-controlled artificial ventilation.
• To propose and evaluate a modified constricted class topper optimization (C-CTO) algorithm for tuning ventilator controllers.
• To improve the dynamic response and stability of artificially ventilated human respiratory systems.

Main Methods:

• A pressure-controlled ventilation (PCV) model was utilized.
• A proportional-integral-derivative (PID) controller was implemented for pressure control.
• Three optimization algorithms—Particle Swarm Optimization (PSO), Class Topper Optimization (CTO), and the proposed C-CTO—were applied to tune the PID controller.

Main Results:

• The performance of swarm-based controllers was evaluated in the PCV system.
• Key performance metrics included settling times and maximum overshoot.
• The C-CTO algorithm demonstrated effectiveness in tuning the PID controller for the artificial ventilator.

Conclusions:

• Swarm-based optimization significantly enhances the dynamic response of pressure-controlled artificial ventilation.
• The proposed C-CTO algorithm offers a viable approach for tuning controllers in artificial respiratory systems.
• The method, validated on a piston-motor lung model, has potential for application in more complex artificial ventilation scenarios.