Liangsong Lu ¹ Yixuan Wang ^2* Guolang Shen ^3* Xie Fei ^4*

• ¹ School of Automation Science and Electrical Engineering, Beihang University, Beijing, Beijing, China
• ² Beihang University, Beijing, China
• ³ Jingdezhen Ceramic University, Jindezhen, China
• ⁴ First Affiliated Hospital of Chinese PLA General Hospital, Beijing, Beijing Municipality, China

Existing Mechanical Insufflation-Exsufflation (MI-E) devices often overlook the impact of cough airflow pressure on mucus clearance, particularly lacking in control over airway pressure during the expiratory phase, which can lead to airway collapse and other types of airway damage. This study optimizes the design of cough assist system and explores the effectiveness of PID and adaptive control methods in regulating airway pressure. The adaptive control method compensates for hose pressure drop by online estimation of the ventilatory hose characteristics. It achieves precise tracking of target pressure and ensures the generation of peak flow rates effective for mucus clearance, even in the absence of known patient lung physiological states and unknown hose leakage parameters. Through a series of comparative experiments, this paper confirms the significant advantages of adaptive control in reducing oscillations and overshoot, capable of more stable and precise airway pressure adjustments. This improved control strategy not only enhances clinical safety but also significantly improves therapeutic outcomes and reduces the risk of complications. The findings indicate that the revamped cough assist system, employing an adaptive control strategy, can effectively prevent airway damage during assisted coughing, offering a safer and more effective sputum clearance solution for critically ill patients with expectoration disorders.