ADVANCED CONTROL OF AMMONIA REACTOR USING BLOCK-ORIENTED MODELS

Abstract

Ammonia is one of the most important chemicals produced as it enjoys the wide use in the manufacture of fertilizers. Hence, modeling and simulation of ammonia manufacturing process has received considerable attention among the process industries way from 1960s. Most chemical processes are inherently nonlinear. In recent years, nonlinear process identification has received more attention in both research institutions and industries. One way to specify model structure is to combine linear dynamic models with static (memoryless) nonlinear functions. These kind of nonlinear models are called Block-Oriented models. Mostly, unless specified otherwise, all studied nonlinear models will have a block-oriented structure: different building blocks will be assembled in specific configurations and studied. Commonly used block-oriented models are Hammerstein model, Weiner model and their combinations. A Hammerstein model has nonlinear block followed by a linear model and a Weiner model has nonlinear model with a linear dynamic block followed by a static nonlinear function. In this work, ammonia reactor model is taken and simplified by using some assumptions and approximations. Then its state-space model is developed. Using the state-space model, the reactor temperature is controlled.