Implementation of Gain Scheduling Multiloop PI Controller Using Optimization Algorithms for a Dual Interacting Conical Tank Process

The design and implementation of a Gain Scheduling Multiloop PI Controller for a Dual Interacting Conical Tank Process are detailed in this paper. Empirical modelling of the process is carried out. The various nonlinear regions in the response are linearized and the state space model and transfer function model is derived for the steady state/equilibrium values. Interaction effect in the various operating regions and the pairing of manipulated variables and outputs are analyzed using Relative Gain Array, Condition Number, and Niederlinski Index. The Gain Scheduling Multiloop PI Controller is then tuned and implemented using optimization algorithms like Genetic Algorithm (GA) and Particle Swarm Optimization Algorithm (PSO). The resulting output specifications are compared and shown in this paper.