Abstract

This paper aims to implement Hybrid PSOGSA tuned Fractional Order (FO) Proportional Integral (PI) controller for the combination of Spherical Conical Interacting Tank Process (SCITP) in real time using Labview. Most of the industrial control systems are complex, multivariable, possess nonlinearity, dead time and there are often complicated interactions between different signals. Taking all the above said properties into consideration this process is considered for study and the objective is to maintain the level of spherical as well as conical tank system. For this complex system, initially Zeigler Nichol‘s (ZN) tuned PI controller is designed and then hybrid PSOGSA optimization technique is used to tune the PI controller parameters. Also FO PI controller is designed and then hybrid PSOGSA algorithm is used to tune the FO PI controller parameters. The hybrid algorithm reduces the possibility of getting trapped in local optima. The servo response and regulatory response are obtained using the optimized values and the results are compared. From the analysis of real time results, it is clear that the proposed hybrid PSOGSA optimized FO PI controller shows improved performance comparatively for SCITP.

Keywords

Evolutionary algorithm. Fractional Order Control, Hybrid PSOGSA, Gravitational Search Algorithm, Two Input Two Output, Particle Swarm Optimization, Spherical Conical Interacting Tank Process.