Control of Blood Glucose Level using Intelligent Techniques

This paper attempts to control the plasma level of glucose for a diabetic patient. A mathematical model of glucose and insulin interactions in the patient's body proposed by Bergman is used in this work. A two-degree of freedom PID control scheme is designed to control the insulin injection in the body for the efficient regulation of plasma glucose level. The controller parameters are optimized using different evolutionary algorithms i.e., ACO, PSO and GA. A conventional PID controller is also used to control plasma glucose level and the response is studied for 24 hours. A disturbance term representing the glucose infused due to meal intake is also included to analyze the control performance during disturbances. It is revealed from the results that GA tuned 2 DOF PID controller provides the best performance among other designed controllers.