Port Hamiltonian Modelling of an Integrated Mechanical Ventilator-Human Respiratory System

Abstract

Mechanical ventilation is a life-saving treatment for critically ill patients who are struggling to breathe independently due to injury or disease. Globally, large numbers of individuals have always required mechanical ventilation per year. In this research work, a compre- hensive integrated Mechanical ventilator human respiratory system which is a complex dynamical system is modelled using the Port Hamiltonian approach. The Port Hamilto- nian framework provides a powerful tool for the analysis, modelling, and control of such systems. The objectives of this research work are separated into four main outcomes: The first objective is formulation of the human respiratory system using a Port Hamiltonian based approach. In this formulation, the first abstraction is a nominal Port Hamiltonian mechanical ventilator human respiratory system model. The second level of abstraction formulates a comprehensive integrated Port Hamiltonian model of the human respiratory system which includes the alveoli. The third abstraction is the Multi-physics and Multi- scale Port Hamiltonian model of the human respiratory system, which is integrated with the mechanical ventilator. The second objective is the formulation of a comprehensive Port- Hamiltonian model of a mixed finite/infinite dimensional electro-hydro-mechanical model of the mechanical ventilator. The third objective is the presentation of a multi-physics Port Hamiltonian generalized model of a Mechanical Ventilator-Human respiratory sys- tem with Naiver stokes for the fluid flow in the mechanical ventilator respiratory system model. The generalized formulation leads to a new class of algebraic constraints in the modelling of the mechanical ventilator respiratory system. The main result presented is a novel interconnected Dirac structure of the fluid/electrically coupled system. Finally, two novel applications of the mechanical ventilator human respiratory system model are presented. Simulations for this research work are conducted using the Matlab platform and the respiratory system model results and the mechanical ventilator results are com- pared to results in existing literature and are found to be comparable. This work can be developed further in future to include robust controllers in the Port Hamiltonian frame- work for improved mechanical ventilator patient dynamics.