Glucose-insulin system dynamics in type-I diabetes patient based on Aquila Optimization-Tilt Integral-Derivative-Filter (AO-TIDF) approach

This study illustrates the conceptual framework of Aquila Optimization-Tilt Integral-Derivative-Filter (AO-TIDF) for managing the recommended insulin dosages to interaction in Type-I Diabetes Mellitus (TIDM) patients using an Artificial Pancreas (AP) for retaining control over Blood Glucose (BG) levels. Using the Aquila Optimization Algorithm (AOA), which regulates the suggested patient model’s controller gains are used to enhance BG control. The purpose of this typical controller (AO-TIDF) is to enhance the robustness and effectiveness of the patient model performance. Glycemic management is hampered by inconsistencies within the patient-focused model. The nonlinear behavior of patient models is efficiently managed by employing an AP-based AO, which also keeps BG levels within the acceptable glycemic range (70–120 mg/dl). After all accuracy, consistency, robustness, noise reduction, and boosted capability to manage uncertainty are assessed by using the recommended patient role models employing AO-TIDF. The explanations behind the recommended approach’s better control capability are shown through a comparison of the outcomes of several control techniques, where it achieved improvements around 20% in system stability, accuracy and robustness. In the future, we aim to integrate a priori-defined performance constraints, robust optimization frameworks, and Lyapunov-based analysis tools to provide theoretical guarantees alongside empirical results. Additionally, we will focus on incorporating a priori robust stability criteria into the controller synthesis phase to ensure guaranteed stability.