Robust Reduced-Order Model Based Postprandial Glucose Regulation in Type-1 Diabetes: An IMC Approach

The treatment of Type 1 diabetes using the artificial pancreas (AP) is considered one of the safety-critical and challenging control problems. The control law should be easily implemented with low complexity and good regulatory performances. In view of this, the present paper proposes a new model-based feedback regulation strategy in which the control-oriented insulin-glucose regulation system is simplified in such a way that the performance of the reduced-order model matches with the actual version. The dominant pole retention-like methodology is applied to obtain the reduced-order model. Now, based on the obtained reduced model, the internal model control (IMC) approach is employed to design a PI controller. Simulation studies have been conducted on the virtual patient and the applied approach produces satisfactory responses in presence of meal ingestion. The applied approach is also robust against model uncertainties.

Automated summary

This paper proposes a new model-based feedback regulation strategy for the treatment of Type 1 diabetes using the artificial pancreas, which shows good regulatory performances and robustness against model uncertainties. A PI controller is designed using the internal model control (IMC) approach, and satisfactory simulation studies are conducted on a virtual patient.