Integrated Pneumatic Servo Module for Mechanical Ventilation With ADRC and RLS-Based Feedforward Control

TY - JOUR

T1 - Integrated Pneumatic Servo Module for Mechanical Ventilation With ADRC and RLS-Based Feedforward Control

AU - Ming, Xiaochuan

AU - Ren, Shuai

AU - Wang, Xiaoyi

AU - Wang, Tao

AU - Lv, Yuying

AU - Luo, Zujin

N1 - Publisher Copyright: © 1982-2012 IEEE.

PY - 2026

Y1 - 2026

N2 - This article presents a miniaturized integrated pneumatic servo module with hybrid adaptive control for mechanical ventilation. Unlike conventional ventilators requiring three or more actuators, the proposed design achieves air-oxygen blending and pressure/flow regulation using only two proportional solenoid valves. The control architecture adopts dual closed-loop control, with parallel active disturbance rejection control for the inner flow loops and PI control for the outer pressure loop. Online recursive least squares identification of respiratory mechanics deployed for adaptive feedforward compensation. Fuzzy control was adopted for positive end-expiratory pressure (PEEP) regulation. Experimental results demonstrate a 300 ms rise time with zero overshoot in pressure tracking, FiO2 control accuracy within 5% across the 30-90% oxygen concentration range, PEEP control accuracy within ±1.5%, and robust performance under varying respiratory mechanics. The integrated design reduces system complexity while maintaining clinical-grade performance, offering a cost-effective solution for resource-constrained healthcare settings.

AB - This article presents a miniaturized integrated pneumatic servo module with hybrid adaptive control for mechanical ventilation. Unlike conventional ventilators requiring three or more actuators, the proposed design achieves air-oxygen blending and pressure/flow regulation using only two proportional solenoid valves. The control architecture adopts dual closed-loop control, with parallel active disturbance rejection control for the inner flow loops and PI control for the outer pressure loop. Online recursive least squares identification of respiratory mechanics deployed for adaptive feedforward compensation. Fuzzy control was adopted for positive end-expiratory pressure (PEEP) regulation. Experimental results demonstrate a 300 ms rise time with zero overshoot in pressure tracking, FiO2 control accuracy within 5% across the 30-90% oxygen concentration range, PEEP control accuracy within ±1.5%, and robust performance under varying respiratory mechanics. The integrated design reduces system complexity while maintaining clinical-grade performance, offering a cost-effective solution for resource-constrained healthcare settings.

KW - Active disturbance rejection control (ADRC)

KW - adaptive control

KW - mechanical ventilation

KW - pneumatic servo systems

KW - recursive least squares (RLS)

UR - https://www.scopus.com/pages/publications/105035504896

U2 - 10.1109/TIE.2026.3675040

DO - 10.1109/TIE.2026.3675040

M3 - Article

AN - SCOPUS:105035504896

SN - 0278-0046

VL - 73

SP - 12691

EP - 12701

JO - IEEE Transactions on Industrial Electronics

JF - IEEE Transactions on Industrial Electronics

IS - 9

ER -