Title: Performance monitoring and analysis of level control using soft computing in a nonlinear spherical tank system

Authors: Gajanan M. Malwatkar; Maheshkumar S. Patil

Addresses: Department of Instrumentation Engineering, Government College of Engineering, Jalgaon, Maharashtra, India ' Department of Instrumentation Engineering, Government College of Engineering, Jalgaon, Maharashtra, India

Abstract: Optimising controller settings for nonlinear unstable process models is examined through both conventional methods and reinforcement learning (RL) algorithms. The effectiveness of the proposed approach is evaluated through a comparative analysis with classical controller tuning techniques. A real-time system is utilised to derive the mathematical model via experimental study, which serves as the basis for the investigation. The suggested approach is implemented on a MATLAB package used for a nonlinear spherical tank process. The results suggest that the RL algorithm demonstrates commendable performance on the models of nonlinear unstable processes examined within this investigation. Specifically, the reinforcement learning tuned proportional integral controller demonstrates improved process characteristics, including excellent time domain specifications, disturbance rejection, achieving seamless reference tracking while enhanced fault-tolerant control, compared to other controller tuning methods. The simulation results underscore the effectiveness and practical applications of the proposed algorithm.

Keywords: disturbance rejection; experimentation; modelling; performance analysis; reinforcement learning; spherical tank.

DOI: 10.1504/IJSCC.2025.145790

International Journal of Systems, Control and Communications, 2025 Vol.16 No.2, pp.119 - 132

Received: 25 May 2024
Accepted: 11 Dec 2024

Published online: 23 Apr 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article