Maximum Sensitivity-Based PI Controller Design for First-Order Unstable and Integrating Systems

Abstract

The proposed paper focuses on introducing a simple and practical PI control approach based on the direct synthesis method for unstable first-order or integrating processes with time delay. The control scheme highlights the enhancement on set-point tracking and disturbance rejection so that good performance can be analyzed. A PI controller is used for disturbance rejection, while a set-point filter is added to improve tracking response. The controller parameters are directly related to the process model and tuned using a single adjustment parameter selected through the maximum sensitivity to ensure robustness. By avoiding the derivative term, the proposed controller remains less sensitive to noise and is therefore more suitable for industrial use. Clear tuning guidelines are provided for different Ms values, making the method easy to apply even for processes with large time delays. Simulation results are compared with several unstable and integrating processes, which shows that the proposed approach delivers improved or comparable performance to existing methods, while maintaining the same level of robustness. The suggested approach is also verified for nonlinear continuous stirred tank reactor (CSTR) based on performance metrics like (ISE), (ITAE), (IAE), and (TV), are used to measure how well the controller works.