In this study, a novel and optimized control policy over type 1 diabetes was sought to e xplore. A nonlinear model of type 1 diabetes (Augmented Minimal Model) was taken into consideration, which has been implemented in MATLAB-SIMULINK environment. The regarded model is one of the proposed diabetes models which consider the conditions of patient. There are uncertainties in this model due to factors such as blood glucose concentration, daily meals and sudden stresses. In addition to different approaches toward the elimination of these uncertainties, distinct control policies could be conducted to monitor blood glucose levels. In this article, a meal disturbance is inserted into diabetic model to consider the real environment and eliminate the regarded uncertainties in the simulation. Moreover, fuzzy control theory was utilized as a logical tool that is used to transform words into actions. To enhance the system performance, a fuzzy type-2 algorithm with independent coefficient had been implemented. Finally, fuzzy type-1 and fuzzy type-2 controllers were compared. It was concluded fuzzy type-2, had a better performance compared to fuzzy type-1. Blood glucose level experienced a lower maximum level in fuzzy type-2. Moreover, it was clarified the controller performance improves considerably by conducting fuzzy type-2 and the controller maintained blood glucose level in the desired range better in comparison to fuzzy type-1.

Keywords