Abstract:
A new combination of fractional order (FO) nonlinear control and sliding mode observer (SMO) for blood glucose regulation in type 1 diabetes mellitus is proposed in this paper. An observer estimates the difficult-to-measure process variables that are significant to control law design and to prevent failures. A SMO is proposed to estimate non-measurable states variables of Bergman minimal model by data acquired from a continuous glucose monitoring sensor. At first based on Backstepping method with FO sliding surface a control signal is proposed, and then interval type 2 fuzzy logic is utilized to reduce the chattering phenomenon in control signal. The FO sliding mode control provides robust performance and the Backstepping algorithm protects the controller in front of mismatched and matched uncertainties. Simulation results of the proposed controller are compared with its integer counterpart. © 2018, Springer-Verlag GmbH Germany, part of Springer Nature.