A nonlinear active disturbance rejection control strategy for blood sugar regulation in type 1 diabetic patients

Type 1 diabetes mellitus (T1DM) is a chronic autoimmune disease that requires lifelong insulin replacement therapy for affected individuals. Traditional model-based control strategies often yield poor closed-loop blood glucose regulation in the presence of disturbances and uncertainties. This work proposes a nonlinear reduced order unified cascade extended state observer (NRUCESO)-based control strategy for regulating the level of blood glucose concentration (BGC) in patients suffering from T1DM. Optimal values of the associated tuning parameters are obtained using crayfish optimisation algorithm (COA) to achieve improved disturbance rejection and tracking performance. Stability of the proposed controller is demonstrated using describing function and Nyquist plots. A comparative analysis is conducted between NRUCESO and three different control schemes which demonstrates the superiority of the proposed controller. Robustness of NRUCESO is illustrated by perturbing various parameters of the nonlinear Bergman's minimal model (BMM) by ±30%, highlighting its ability to maintain closed-loop system stability and provide satisfactory performance under varying conditions of the patient. A real-time controller-hardware-in-loop (C-HIL) is utilised to verify the performance of various controllers.