Optimal Intelligent Control for Glucose Regulation

This paper introduces a novel control methodology based on fuzzy controller for a glucose-insulin regulatory system of type I diabetes patient. First, in order to incorporate knowledge about patient treatment, a fuzzy logic controller is employed for regulating the gains of the basis Proportional-Integral (PI) as a self-tuning controller. Then, to overcome the key drawback of fuzzy logic controller, i.e., the lack of systematic methods to define fuzzy rules and fuzzy membership functions, fuzzy PI controller are optimised by Particle Swarm Optimization with Linearly Decreasing Weight (LDW-PSO) algorithm, which is a novel evolutionary computation technique. Simulation results show the effectiveness of the proposed optimal fuzzy PI controller in terms of accuracy and time margin.
This paper introduces a novel control methodology based on fuzzy controller for a glucose-insulin regulatory system of type I diabetes patient. First, in order to incorporate knowledge about patient treatment, a fuzzy logic controller is employed for regulating the gains of the basis Proportional-Integral (PI) as a self-tuning controller. Then, to overcome the key drawback of fuzzy logic controller, i.e., the lack of systematic methods to define fuzzy rules and fuzzy membership functions, fuzzy PI controller are optimised by Particle Swarm Optimization with Linearly Decreasing Weight (LDW-PSO) algorithm, which is a novel evolutionary computation technique. Simulation results show the effectiveness of the proposed optimal fuzzy PI controller in terms of accuracy and time margin