Multiple Model Adaptive Control of Valve Flow Using Event-Triggered Switching

Valve flow control has always attracted research due to the wide breadth of applications it offers. Its use in advanced applications, such as air and oxygen flow control for neonatal ventilation has demanded higher specifications for accuracy and performance. In this paper, we propose a model reference adaptive control scheme for valve flow control of ventilators which utilizes multiple models that correspond to different operating points of the valve. Switching between models operates according to level of flow using a smoothing mechanism. In addition, a inverse hysteresis model is used to attenuate the hysteresis effect. The scheme has been implemented and tested on multiple valves and using several accuracy and stress tests, providing excellent results.