An Analytical Study of Artificial Intelligence Algorithms Used PID Controller to Control Blood Glucose in Diabetics

Abstract

Diabetes is a widespread disease in the world today and research is still underway in the field of artificial pancreas, also known as closed- loop insulin therapy, where it is possible to calculate the rate of insulin infusion that will maintain the concentration of glucose in the blood.

In this paper, four different control algorithms have been proposed that adjust the parameters of the artificial pancreas based on the PID controller. These algorithms are the Genetic Algorithm for Controller Tuning (GA_PID), the Annealing Simulation Algorithm (SA_PID), the Bird Flock Algorithm (PSO_PID) and the Bee Algorithm (PSO_PID). ABC_PID). The study was carried out in a hypothetical diabetic patient based on a modified Bergmann model of glucose and insulin regulation. The simulation was carried out under daily food conditions of three meals.

In practical application, the results were promising by regulating blood glucose using the previous algorithms, where the simulated annealing algorithm gave the best time while preserving the permissible values ​​for the concentration of glucose in the blood. This research can be developed by combining two or more algorithms to take advantage of The advantages of each algorithm.