An adaptive control algorithm for blood glucose regulation in Type-I Diabetes Mellitus patients
Akshaya Kumar Patra, Girija Sankar Panigrahi
Abstract
This study presents the design and development of a Remora Optimization -Tilt Integral-Derivative-Filter (RO-TIDF) controller to infuse ideal insulin doses to manage Blood Glucose (BG) levels in Type-I Diabetes Mellitus (TIDM) patients with Artificial Pancreas (AP). In this control technique, Remora Optimization Algorithm (ROA) is used, which adjusts the controller gains of the proposed patient model for better regulation of BG. This classic RO controller is designed to improve patient performance and robustness. Nonlinearities present in the patient model pose problems for glycemic control. Using an AP-based ROA efficiently overcomes the nonlinearity of patient models and maintains BG levels in the glycemic band (70–120 mg/dl). The suggested patient model with RO-TIDF is examined for precision, consistency, robustness, noise reduction, and improved uncertainty handling. A comparative study of results from different control methods reveals the causes of the advanced control performance of the proposed approach.