An Automatic Artificial Pancreas (AP) based on MO-PID Control Algorithm

The goal of the article is the design of a Mayfly Optimization-PID Controller (MO-PIDC) to inject an ideal dose of insulin via AP for control of BG level in the Type-I Diabetes Mellitus (TIDM) patients. This classical controller with optimization technique is seen to improve performance and to make the system robust. Robustness is to be improved by considering control parameter as a function of error. This is utilized in the proposed novel control technique. The nonlinearities present in TIDM patient framework cause BG control issues. The use of AP based Mayfly optimization algorithm efficiently to maintain BG level in the range of normo-glycemia. This proposed patient model with MO-PIDC is analyzed for improved accuracy and stability, enhanced robustness, noise suppression, and better ability to handle uncertainties.