Abstract

This paper presents a mathematical model and simulation of an artificial respiratory system designed to mimic natural breathing behavior. The model incorporates key respiratory mechanics, including airway resistance and lung compliance, to accurately reproduce patient breathing dynamics. First, the ideal response of the respiratory system is evaluated as a reference baseline. Then, system stability and robustness against disturbances in airflow and pressure are enhanced using both classical PID and Fractional-Order PID (FOPID) controllers. The controller parameters are optimally tuned using two evolutionary optimization techniques: Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). Comparative results indicate that PSO is more efficient for tuning conventional PID controllers, yielding lower error indices and better variance explanation. In contrast, GA provides superior performance when optimizing FOPID controllers, benefiting from the added flexibility of fractional-order terms. Overall, FOPID controllers outperform standard PID, demonstrating strong potential for advanced respiratory therapy applications.

About the Speaker

Mr. Osaid Mohammed is a Graduate Student in the Control and Instrumentation Engineering Department at King Fahd University of Petroleum and Minerals (KFUPM), Dhahran, Saudi Arabia. He earned his Bachelor's degree in Mechatronics Engineering from Taiz University, Yemen. His research focuses on safe autonomous systems.