Model Identification of Two Double-Acting Pistons Pump: A NARX Network Approach

The demand for utilizing medical ventilators for inter and intrahospital transport has risen quickly in recent periods of human society due to the consequences of the Covid-19 pandemic. Patients who have suffered during this pandemic need different modes of ventilation as a vital cure for their injured lungs. The development of control modes for medical ventilators requires a mathematical model which is able to represent the behavior of an air compressor accurately. In this paper, a parametric black-box method is used to obtain such a mathematical model, that is Nonlinear AutoRegressive with eXogenous input (NARX) model. This NARX model is constructed by experimental data from the air compressor using the bump-based two double-acting pistons (DAPP). The term selection and parameter estimation process are presented and conducted based on measured data. The identification model is also validated based on statistic criteria in the aspect of one-step-ahead prediction (OSA). The results show great tracking performances of identified NARX model.