Closed Loop Experimental Validation of Linear Parameter Varying Model with Adaptive PI Controller for Conical Tank System

Abstract

The control of the liquid level in a conical tank system is a complicated task in many process industries, where it is used for drainage of slurries, viscous liquids and solid mixtures. The conical tank system is highly nonlinear due to its variation of the area of cross section, with respect to height of the system. The process identification of this nonlinear system is done, using the linear parameter varying method. Linear Parameter Varying (LPV) modeling is capable of describing the system over its entire operating trajectory. Based on the open loop response of the conical tank, the entire region of the nonlinear process is split into approximate linear regions, and their respective transfer function models are formed. Then the linear parameter varying model is identified, by interpolating the transfer function models. The obtained model is investigated and validated with the real time process data. Further, a conventional PI controller is designed for the control of the liquid level in the conical tank system, using the direct synthesis method. To improve the closed loop performance, the adaptive PI control scheme is designed and implemented for the conical tank system. The single PI controller and Adaptive PI controller are designed for the LPV model, and it is implemented on the conical tank system in real time. From the experimental results, it is proved that the performance of the Linear Parameter Varying model based PI and the Adaptive PI controller is significantly better than that of the other control strategies.