Performance Investigation of EMFO-Based Perpetual Online Tuned Variable SR-Controller in Simulated Real Environment

Abstract

Abstract This study uses the enhanced moth flame optimization (EMFO) algorithm with an online tuning approach to optimize the parameters of a variable range split range PID (SRPID) controller. Before implementing the developed EMFO algorithm in the actual plant, it is necessary to investigate the performance of the same in the simulated real environment. Therefore, in the present work, an electrical analogous model of the practical environment is simulated for investigation by considering several effects, namely imperfect insulation, density, viscosity, and compressibility. Further, to check the effectiveness of the proposed algorithm, the controller performance using the EMFO algorithm is compared with the performance using the original MFO algorithm. The validation results show a substantial improvement in the case of EMFO-based controller with an online tuning method in comparison to MFO-based controller. EMFO algorithm exhibits superior performance as it combines the benefits of three modifications (change the spiral path, opposition learning-based initialization, and change in flames selection) in the original MFO algorithm. Furthermore, the system is also investigated for the effect of system dynamics and process disturbance. It is concluded that the developed EMFO algorithm gives superior performance in a simulated real environment paving the way for possible implementation in practical situations.