Chapter 10 - Applications of Identification in Process Control

Abstract

This chapter focuses on applications of identification in process control. It describes an industrial control project approach and the use of identification in various steps in the control project. The chapter reviews the characteristics of industrial processes and their identification requirements by the processes of the refinery/petrochemical industry. Proportional, integral, and differential (PID) tuning is also part of the pre-test in a model predictive control (MPC) project. Therefore, proper tuning of PID loops is very important to maintain good performance of the overall process control system. The chapter outlines the model-based PID autotuning method. The chapter describes various ways to improve the estimate of the low-gain direction of the model and the influence of model structures, different identification methods, and test methods. The purpose of this chapter is twofold: (1) to provide guidelines for industrial process control and identification and (2) to show that the modern identification approach is very cost-effective.