LPV Modeling and Performance Analysis of Boiler Drum Using IMC-PI Controller

Abstract

This paper presents the development of linear parameter varying model and control strategy for Industrial boiler drum. The Boiler drum is modeled using the fundamental mass and energy balance equations and data collected from the real time plant. Based on the open loop response of the Boiler drum, the whole non-linear process is split into 4 approximate linear regions and their respective transfer function models are formed from the input-output data. Since the process is nonlinear in nature, linear transfer function models are developed in each operating regions and they are interpolated to obtain the linear parameter varying model. The LPV model is validated using the real time data. Since a single PI controller for whole process does not satisfy both servo and regulatory performance, a multi-model approach with four PI controllers is implemented for the process. Internal Model Control ( IMC)-PI controller is designed to analysis the servo and regulatory performance of the developed LPV model.